BOARD CHARACTERISTICS AND EARNINGS MANAGEMENT OF LISTED FOODS

AND BEVERAGES FIRMS IN NIGERIA

IDRIS IBRAHIM
(MSc/ADMIN/2401/2011-2012)

BEING A DISSERTATION SUBMITTED TO SCHOOL OF POSTGRADUATE

STUDIES, AHMADU BELLO UNIVERSITY, ZARIA IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE AWARD OF MASTER OF SCIENCE DEGREE IN
ACCOUNTING AND FINANCE

DEPARTMENT OF ACCOUNTING
FACULTY OF ADMINISTRATION

DECEMBER, 2015

i
 Declaration
I hereby declare that this dissertation titled „Board Characteristics and Earnings Management of

listed foods and beverages firms in Nigeria‟ is original and has been written by me under the

supervision of Prof. Musa Inuwa Fodio and Mallam Lawal Muhammad. To the best of my

knowledge and belief, this work has never been submitted or presented to any institution for the

award of degree or diploma. In addition, the materials consulted and used in the study have been

duly acknowledged in the text and by means of references.

_____________ ____________
Idris Ibrahim Date

ii
 Certification

This dissertation titled „Board Characteristics and Earnings Management of listed foods and

beverages firms in Nigeria‟ meets the regulations governing the award of the degree of Master of

Science (M.Sc.) in Accounting and Finance of Ahmadu Bello University Zaria; and therefore

approved for its contribution to knowledge and literary presentation.

____________________ ______________
Prof. Musa Inuwa Fodio Date
Chairman, Supervisory Committee

____________________ ______________
Mallam Lawal Muhammad Date
Member, Supervisory Committee

____________________ ______________
Dr. A. B. Dogarawa Date
Head of Department

____________________ ______________
Prof. Kabir Bala Date
Dean, Postgraduate School

iii
 Dedication

This research work is dedicated to Allah (SWT), the most Beneficent, most Merciful; and to our

beloved prophet Muhammad, peace and blessing of Allah be upon Him. I thank Allah for giving

me the opportunity to complete this research work successfully.

iv
 Acknowledgements

All praise is due to Allah. May Allah's Peace and Blessings be upon His final Messenger, his

pure Family, his noble Companions, and all those who follow them with righteousness until the

Day of Judgment. I thank Almighty Allah (who has power over everything) for giving me the

opportunity to bring this work to its logical conclusion.

I would like to gratefully and sincerely thank my major supervisor, Prof. Musa Inuwa Fodio, for

his guidance, understanding, patience, brilliant advice and constructive criticisms as he

relentlessly read this Dissertation. I would also like to extend my sincere gratitude to my second

supervisor, Mallam Lawal Muhammad, for his inputs in the research work and for encouraging

me to be a good academic researcher. May Allah in his infinite mercy grant them paradise.

My appreciation equally goes to Dr. Shehu Usman Hassan for his support to my M.Sc. and

research, for his patience, motivation, enthusiasm, and immense knowledge. His guidance in and

out of the class helped me in writing this Dissertation. May Allah continue to guide him in all his

affairs.

I would also like to thank the department of accounting, especially those members of academic

staff who made their input to this research work in terms of valuable discussion, advice, and

constructive criticisms. In particular, I would like to thank Dr. A. B. Dogarawa (H. O. D.), Dr.

M. S. Tijjani (M.Sc. Coordinator) and Dr. Ahmed Bello for their concern, guidance, and

encouragement toward making my dream comes true. May Allah reward them abundantly.

v
My principal indebtedness is to my parents, Alhaji Idris Sulaiman, Malama Hauwa‟u Jibril and

Malama A‟ishatu for their encouragement and above all, for their numerous and valuable prayers

and advice on how to shape my life; most especially my mother (Hauwa‟u) whose courage and

insight has propelled my quest for a university education. Your reward is with Allah. I pray, for

Allah to make paradise your final destination. My unreserved gratitude and appreciation also

goes to my beloved wife Sakinah for her support, encouragement, understanding, and quiet

patience. Her tolerance of my occasional vulgar moods is a testament in itself of her unyielding

devotion and love. I cannot finish this journey without expressing words of thank to my brothers

and sisters at home. May you all be blessed by Allah.

Lastly, but not the least, I would like to thank my friend and also a colleague, Hussaini Bala, for

providing me with unfailing advice and support and continuous encouragement throughout my

M.Sc. programme and through the process of writing this Dissertation. May Allah reward you

with Aljannah. Also, my gratitude and appreciation go to the following friends: Kabiru Adamu,

Ayuba Abubakar, Abdullahi Abubakar, Mallam Alhassan, Abubakar Ja‟e, Murtala Abdullahi

Dogarawa, and Abba for their concern and good advice.

I must admit that the names mentioned above are far from complete. In fact, the making of a

comprehensive list may necessitate another volume of this size if not more. Since this is not

feasible, I duly thank all those whose names are worth mentioning but because of space

constraint have not been mentioned.

vi
 Abstract
As a response to some financial scandals and corporate failures in Nigeria and around the globe
which are linked to earnings management, certain characteristics of Board of Directors that can
improve their monitoring function are suggested in the literature as corporate governance
mechanisms. Thus, the study concentrated on three board characteristics’ proxies, namely:
Board Competency, frequency of Board Meetings and Gender Mix and their relationships with
earnings management (because, they have not yet been studied extensively in Nigeria).
Therefore, the study investigated the impacts of Board Competency, frequency of Board
Meetings and Gender Mix on Earnings Management (in the context of agency relation) of listed
foods and beverages firms in Nigeria from 2007 to 2013. The estimation of discretionary
accruals (proxy for Earnings Management) is by using modified Jones (1991) model. The sample
size of the population is nine (9) firms. Both correlational and ex-post factors research design
were used. A multiple regression technique was employed to determine the impact of Board
Characteristics on Earnings Management. The result was interpreted using fixed effect- least
square dummy variables. The results reveal that Board Competency has no significant impact on
Earnings Management. The impact of frequency of Board Meetings and Gender Mix on Earnings
Management were however found to be negative and statistically significant. The study
concluded that increase in number of board meetings and the proportion of women directors in
the board constrain the level of discretionary accruals; while directors’ knowledge of accounting
and/or finance (board competency) does not guarantee quality of earnings. Therefore, in line
with the findings and conclusions, the study recommends that SEC should encourage adherence
to at least the minimum requirement of board meetings (four times in a financial year) by making
it mandatory. Government in collaboration with Corporate Affairs Commission should come up
with a policy where by companies will be forced to provide seat for women in their boards, give
them responsibilities in area of finance and control related matters; as this would enhance firm
performance and constrain earnings management, since they normally develop trust leadership.

vii
 TABLE OF CONTENTS

Title- - - - - - - - - - - - -i

Declaration- - - - - - - - - - - -ii

Certification- - - - - - - - - - - -iii

Dedication- - - - - - - - - - - -iv

Acknowledgements- - - - - - - - - - -v

Abstract- - - - - - - - - - - -vii

Table of Contents- - - - - - - - - - -viii

List of Tables- - - - - - - - - - - -xi

List of Appendices- - - - - - - - - - -xii

CHAPTER ONE: INTRODUCTION

1.1 Background to the Study- - - - - - - - - -1

1.2 Statement of the Problem - - - - - - - - -6

1.3 Aims and Objectives of the Study- - - - - - - -10

1.4 Hypotheses of the Study- - - - - - - - - -11

1.5 Scope of the Study- - - - - - - - - -11

1.6 Significance of the Study - - - - - - - - -12

CHAPTER TWO: LITERATURE REVIEW

2.1 Introduction- - - - - - - - - - -14

2.2 Conceptualization- - - - - - - - - -14

2.3 Methods of Earnings Management- - - - - - - -21

2.4 Concept of Corporate Governance, Board of Directors and Agency Problems- - -31

2.5 Nigerian Code of Corporate Governance and Board Characteristics- - - -36

viii
2.6 Review of Empirical Studies- - - - - - - - -37

2.7 Theoretical Framework - - - - - - - - - -50

CHAPTER THREE: STATEMENT OF METHODOLOGY

3.1 Introduction- - - - - - - - - - -52

3.2 Research Design- - - - - - - - - - -52

3.3 Population of the Study- - - - - - - - - -53

3.4 Sample Size and Sampling Technique- - - - - - - -54

3.5 Method of Data Collection- - - - - - - - -54

3.6 Techniques of Data Analysis- - - - - - - - -55

3.7 Model Specification- - - - - - - - - -55

3.8 Variable Measurement-Dependent and Independent Variables - - - -57

3.9 Control Variables- - - - - - - - - - -59

3.10 Justification of the Methodology - - - - - - -60

CHAPTER FOUR: DATA PRESENTATION, ANALYSIS, AND INTERPRETATION

4.1 Introduction- - - - - - - - - - -62

4.2 Descriptive Statistics- - - - - - - - - -62

4.3 Correlation Matrix- - - - - - - - - -65

4.3 Robustness Test- - - - - - - - - - -67

4.5 Regression Results- - - - - - - - - -70

4.6 Test of Hypotheses- - - - - - - - - -70

CHAPTER FIVE: SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

5.1 Summary- - - - - - - - - - - -75

ix
5.2 Conclusions- - - - - - - - - -76

5.3 Recommendations- - - - - - - - - -78

5.4 Limitations of the Study- - - - - - - - - - 79

5.5 Suggestions for Further Research- - - - - - - -80

References- -- - - - - - - - - -82

Appendix - - - - - - - - - -98

x
 LIST OF TABLES

Table 4.1 Descriptive Statistics- - - - - - - - -60

Table 4.2 Correlation Matrix- - - - - - - - -65

Table 4.3 Multicollinearity test- - - - - - - - -67

Table 4.4 Heteroskedasticity Test - - - - - - - -68

Table 4.5 Hausman Test- - - - - - - - - -68

Table 4.6 Regression Results-- - - - - - - - -70

xi
 LIST OF APPENDICES

Appendix I: Listed foods and beverages firms (both population and sample size) - - -98

Appendix II: Data for the study- - - - - - - - -99

Appendix III: Descriptive, correlation and regression results and robustness test - -103

Appendix IV: Fixed Effect-LSDV Regression- - - - - - -106

xii
 CHAPTER ONE
INTRODUCTION

1.1 Background to the Study

Earning is one of the most important items in financial statements. This is because, users of

financial statements mostly focus on the company‟s earnings before looking at other variables.

Earnings represent the image of a company on the eyes of many investors and other financial

statements‟ users for decision-making purposes. Earnings indicate the extent to which a company

has engaged in value added activities. Therefore, increase in earnings represents an increase in

company‟s value, while decrease in earnings signals a decrease in that value (Lev, 1989).

Accounting deals with measurement and communication of economic information that involves

the determination of net income (accounting earnings). Accounting‟s earnings serve as a major

constituent of corporate information required in the capital market for assessing firm

performance and for stock valuation (Musa, Ibikunle & Oba, 2013). Therefore, accounting

earnings‟ information need to be more reliable. This is because, the integrity of financial reports

depends on the reliability of earnings being reported by firms; and the capital market needs

precise and unbiased financial reporting to value securities and revive investors‟ confidence

(Roodposhti & Chasmi, 2011).

The only source through which information is passed from the principal to the owners is through

financial statement. A reliable financial statement is expected to provide vital information to

investors to enable them make the right business decisions. A reliable financial statement is

assumed to provide information free from errors and bias that would enable users to make

accurate judgment regarding the information (Shehu, 2013). Moreover, the responsibility for

preparing and publishing external information lies with the firm‟s managers. As such, managers

1
use their knowledge of the firm and the current state of business circumstance to prepare

information that gives a true and fair view of the firm‟s financial state and performance.

However, due to information asymmetry, managers may use their own discretion in preparing

and reporting financial statement to their own advantage (Scott, 2003). This may give rise to

agency problem.

Agency problem is said to have existed when managers fail to act in the best interest of the

owners. The existence of agency problem results from separation between ownership and

control; as managers would have more inside information than the financial providers

(shareholders). Evidence from literature reveals that managers use their discretion over

accounting numbers to achieve private gain; and flexibility of accounting standard usually gives

room for them to adjust earnings through managing accruals. Managers have many incentives to

manage earnings like compensation, avoid debt covenant violation, meeting and beating

benchmark, reducing regulatory or political cost, to meet analysts‟ expectations and to make a

firm appears a less risky investment (Kasznik, 1999 and Trueman & Titman, 1988). Bunamin,

Abdulra‟uf, Johari and Abdulrahman (2012) have the notion that Earnings Management with the

intention to manage users‟ perception in firms are considered unethical even if no accounting

standards are violated. Hence, Earnings Management has the propensity to mislead which may

be difficult to detect by ordinary people who do not have requisite knowledge on the issue

relating to accounting numbers.

Earnings manipulation makes financial reporting to be of less quality and reduces the level of

confidence of investors in their decision making process (Shehu & Abubakar, 2012). Nowadays,

most users of financial statement do not count accounting earnings as a major yardstick for
2
performance evaluation as well as for decision-making. Evidences from literature and financial

scandals around the globe prove that Earnings Management reduces investors‟ confidence. On

the other hand, Board of Directors are regarded as an important internal corporate mechanism

responsible for mitigating agency conflicts between managers and shareholders by helping in

constraining the level of Earnings Management. Therefore, one of the major roles of Board of

Directors is to monitor and reduce the incidence of Earnings Management (Hashim, Ariff &

Salleh, 2013). They are responsible for monitoring managers on behalf of shareholders and

overseeing the financial reporting processes.

Composing Board of Directors with diverse knowledge, skills, gender, and function is an

important determinant of effective board for carrying out its monitoring function. Board

composition includes the determination of the proportion of independent and executive directors,

mix of qualification and expertise, designating audit, proportion of female directors on the board

and number of Board Meetings (Man, 2012; Bertrand & Mullainathan, 2001). Several studies

suggest that both the informativeness of reported earnings and firm‟s performance are affected

by Board of Directors‟ attributes such as the board size, board independence, frequency of Board

Meetings, Board of Directors‟ competencies and managerial ownership (Vafeas 1999; Klein

2012; Hermalin & Wesberch 2012). Therefore, they are expected to monitor and control the

behaviours of managers to ensure they act based on the shareholders‟ interest. Although, a lot of

literature suggest that effective board helps reduce Earnings Management, but issues related to

corporate governance and Earnings Management are inconclusive. This is because many

corporates failure witnessed around the globe occurred in the developed countries, where they

have more sophisticated, and sound corporate governance system. And the Board of Directors

were mostly held responsible for failure to control the activities of managers.
3
Accounting scandals and large businesses failure such as Pamalat (2003) in Italy, Enron (2000),

Xerox (2001), Worldcom (2000), Satya Computer service in India among others indicate Board

of Directors as internal corporate governance mechanism, might not be able to withhold

managers from Earnings Management behaviours. And many people considered Board of

Directors as passive entities controlled by management. This accounting scandals and businesses

failure around the globe have shaken the integrity of accounting information and resulted in a

drop of investors‟ confidence in capital market. Enron‟s case made the United States congress to

respond by passing Sarbanes Oxley Act of 2002 to remedy perceived deficiencies in financial

reporting. The role of Board of Directors was strengthened; and certain characteristics are

suggested within the corporate governance (Saleh, Iskandar & Rahmat, 2005).

In view of the centrality of Board of Directors in the entrenchment of good corporate governance

practice, many corporate governance codes were introduced in Nigeria (such as the Code of Best

Practices on Corporate Governance in Nigeria- 2003 SEC Code; Code of Corporate Governance

in Nigeria- 2011 SEC Code; Code of Corporate Governance for Banks in Nigeria Post-

Consolidation- 2006 CBN Code among others) to tackle the issue of poor representation by the

management that resulted in failure and bankruptcy of some companies such as Cadbury Nigeria,

African Petroleum (AP), Savannah bank, and the financial scandal of the five banks‟ managers,

all in Nigeria. Therefore, to improve the monitoring functions of board of directors, most of the

commendable features of the 2011 SEC code are boards‟ related. As such, several characteristics

of board of directors are suggested by the code; furthermore, there are also some attributes of

Board of Directors that were not mention by the code, but suggested in the literature to improve

4
the monitoring functions of corporate boards. Several studies suggest that there are some Board

of Directors‟ attributes that have significant impact on Earnings Management.

Considering the fact that most of the businesses failure and financial scandals around the globe

happened in the developed countries where they have more sophisticated corporate governance

system; and the list of recent cases of creative accounting practices seems to be growing as many

corporate bodies in Nigeria are being investigated (Akenbor & Ibanichuku, 2012). Furthermore,

foods and beverages sector of the economy has also witnessed the problem of earnings

manipulation. For example, Cadbury Nigeria Plc. manipulated its financial statement to boost its

image as well as push its stock price. As a result, the company inflated its turnover, profit and

other performance indices. The company had been doing so since 2002 but was later discovered

in 2006 (Onyenweau, 2009). Therefore, the motive behind this study is the implicit assertion

made by Klein (2006) that poor corporate governance and Earnings Management are positively

related; and that most corporate governance codes introduced in Nigeria are designed to

strengthen boards‟ monitoring functions; yet there are still cases of corporate governance

malfunctions. Several bodies of literature have identified various board attributes that can

improve their monitoring functions. However, up to date there is no consensus as to what

combination of board of directors‟ characteristics constitutes effective board in monitoring

managerial incentive to manage earnings.

The study is aim at determine the impacts of board characteristics on earnings management.

Three Board of Directors‟ characteristics of listed foods and beverages firms in Nigeria will be

considered as the independent variables, namely: Board Competency, frequency of Board

Meetings and Gender Mix. While the dependent variable will be Earnings Management, proxy

by the discretionary accruals of listed foods and beverages firms in Nigeria.

5
1.2 Statement of the Problem

Previous literature such as Hashim & Devi (2008), Buamin et al. (2012) and Gulzar & Wang

(2011) recognize Earnings Management as a potential agency cost, since managers manipulate

earnings to mislead shareholders and fulfil their own interest. Therefore, the Board of Directors

that is in charge of minimizing the agency conflict between managers and shareholders are

expected to play a significant role in constraining the level of Earnings Management (Lei, 2014).

Hence, most of the literatures that examined the relationship between Earnings Management and

Board Characteristics suggest that Earnings Management erode the quality of financial

statement. It also suggests that Board of Directors are not performing their duty effectively to

constraint the level of Earnings Management.

High profile of accounting scandals and business failure around the globe (most especially in the

developed economies) from the beginning of the 21st century highlighted the intentional

misconduct of managers towards reporting earnings. Failure of large businesses such as Enron

(200), Xerox, Worldcom (2000), Aldephia, Tyco, Parmalat, One-tel, Savannah bank, and

Cadbury Nigeria Plc. raised concerns over the weakness of corporate governance/Board of

Directors as it failed to protect the interests of shareholders; and the reliability on financial

statement by shareholders for decision making became questionable (Musa, et al. 2013). This

drew the attention of corporate governance reform around the globe. Consequently, the Sarbanes

Oxley Act was introduced in 2000 by the United States to tackle the situation.

6
According to CBN (2006), a survey by SEC indicates that weak corporate governance accounted

for the corporate failures in Nigeria, as only 40% of the listed companies recognized the code of

corporate governance in place The list of recent cases of creative accounting practices seems to

be growing as many more corporate bodies in Nigeria are still being investigated (Akenbor &

Ibanichuku, 2012). The collapse of African International Bank, Savannah Bank, Cadbury

Nigeria, the account manipulation of African Petroleum (AP) and the sack of five banks‟

managers have all been linked to Earnings Management in Nigerian public companies (Odia &

Ogiedu, 2013). Previous studies recognise board as a crucial internal control mechanism and

suggest that an effective board helps in mitigating Earnings Management activities.

Despite the fact that there are many empirical studies that investigated the issue of Board of

Directors‟ attributes/corporate governance and Earnings Management, majority focused on

determining the statistical impact of board size, board independence, CEO duality, managerial

ownership and audit committee on Earnings Management. These prior studies includes: Saleh,

Ja‟afar & Hassan (2008), Shehu & Mamman (2011), Moustapa & Ragab (2011), Uadiale (2012),

Odia & Ogiedi (2013), Ugbade, Lizan & Kaseri (2013), Musa, et al. (2013), Shehu (2013), Al-

Damari & Ismail (2013), Swastika (2013), among others. However, Board Competency,

frequency of Board Meetings and Gender Mix have not yet been studied extensively.

Board competency is a nomenclature used to refer to board of directors‟ financial expertise in

relation to earnings management. In this study focus was made on board of directors‟ financial

background. According to Yousef, Nur Hidayah, & Khairil Faizal (2014) and Ebraheem &

Mohamad (2012), to monitor the financial reporting process, control manipulation, make

information more transparent and address financial statements‟ issues, the directors must have
7
accounting knowledge. Empirical studies show that financial expertise is an important

determinant of quality financial statement. Kantudu and Ishaq (2015) assert that board members

who do not have financial and/or accounting knowledge cannot detect earnings management; and

their inability to detect potential earnings manipulation by the management in financial statement

will make them mere spectators in the financial reporting process. Meanwhile, literature supports

the idea that Board of Directors‟ competency improves the quality of financial reports.

Nevertheless, the few prior empirical researches on the effect of Board Competency on Earnings

Management documented different statistical result. The study focused on investigating the

influence of board competency on earnings management of listed foods and beverages firms in

Nigeria; considering the fact that these companies have on average 30% competent directors

with maximum and minimum of 7% and 50% board competency. When compared with other

studies that used dichotomous variables to represent present or absent of board competency in

the team of board directors, is an indication that not all the companies abroad have at least one

board member that is competent. Despite there are competent directors in these companies,

earnings management have been noticing. Therefore, the study test to see how board competency

affect earnings management in Nigeria with reference to foods and beverages firms in Nigeria.

More so, the level of interaction and activities has influence on earnings management. Frequency

of Board Meetings is presumed to be a good proxy for the corporate governance to control

managers‟ behaviour. Board that meets frequently are expected to solve the problem effectively.

Effective board is expected to meet regularly to stay on top of accounting and control related

matter to make sure financial reporting process is functioning properly (Zhou & Chen, 2004). On

the contrary, Jensen (1993) argues that most of the Board Meetings are not very effective, since

the board is often forced to engage in high frequency activities to resolve corporate matters.
8
However, empirical researches on Board Meetings and Earnings Management are rare and

mainly from abroad; and have been reporting different direction of association. Some researchers

like Jaiswal & Banerjee (2009), Xie et. al. (2003), Zhou & Chen (2004) and Vafeas (1999)

reported that frequency of Board Meetings is negatively associated with Earnings Management;

Gulzar & Wang (2011) reported that number of Board Meetings have positive association with

Earnings Management. Although, because of the importance of board meetings, Nigerian SEC

code provided that board of directors should at least meet four times every financial year. But

looking at the meetings held by these companies within the period under review, one can attest

that there is a lack of consistency in the number meetings held. Some of them held meetings

below the minimum requirement while some above (up to eight times). This inconsistency may

be due to the flexibility of the code or based on the matter arises at each firms. This may also

affect the magnitude of earnings management of the said firms. Therefore, investigation needs to

be carried out to access the impact of frequency of board meetings on earnings management in

the context of Nigerian foods and beverages firms in order to ascertain position.

According to Gallego, Garcia & Rodriguez (2010), board of diverse gender may better avoid

practice of Earnings Management, hence, provides shareholders with more reliable financial

reporting. Oscar and Daniel (2013) argued that female board member improve board monitoring

and hence prevent earnings management to a larger extent. This is because male are likely view

leadership as a series of transaction with subordinates, while female are more likely to have more

interactive leadership style. Also male leaders are more concerned with making money and

career advancement, while female are more interested in helping people (Krishnan & Pearson,

2008).

9
Despite there is these arguments in favour of women directorship, in reality, their representation

in the foods and beverages firms‟ team of board of directors is very low, as some firms within

the sector did not provide even a single seat for women. Even those that provided the seat for

them their proportion is very low. On the contrary, some researchers like Hili & Affes (2012)

and O‟Reilly and Main (2012) suggest that women directors are only appointed for symbolic

rather than for substantive reasons. This research also focused on the representation of women on

board, with the argument proffered that boards with women are likely to be more effective than

homogeneous boards.

In general, available literature in Nigeria shows that there is dearth of research work on the

influence of Board Competency, Board Meetings, and Gender Mix on Earnings Management.

Hence, this study is aimed at determining the possible relationship that exists between the

various Board Characteristics (Board Competency, frequency of Board Meetings and Gender

Mix) and Earnings Management with reference to listed foods and beverages firms in Nigeria.

These Board Characteristics‟ variables to the best of our knowledge have not been captured

altogether in any research work relating to Earnings Management in Nigeria (that is, based on the

literature reviewed). Hence, the combined effect of these variables on Earnings Management is

hoped to yield an effective result different from other researches.

Therefore, the study tries to provide answers to the following research question raised. „How

does Board Competency affect Earnings Management of listed foods and beverages firms in

Nigeria? To what extent does frequency of Board Meetings affect Earnings Management of

listed foods and beverages firms in Nigeria? How does Gender Mix affect Earnings Management

of listed foods and beverages firms in Nigeria?‟

10
1.3 Aim and Objectives of the Study

The main objective of this study is to determine the impact of Board Characteristics on Earnings

Management of listed foods and beverages firms in Nigeria. However, the study has the

following specific objectives:

i. To examine the effect of Board Competency on Earnings Management of listed foods

and beverages firms in Nigeria.

ii. To determine the impact of frequency Board Meetings on Earnings Management of listed

foods and beverages firms in Nigeria.

iii. To evaluate the effect of Gender Mix on Earnings Management of listed foods and

beverages firms in Nigeria.

1.4 Hypotheses of the Study

As the main objective is set towards ascertaining the impact of Board Characteristics on Earnings

Management of listed foods and beverages firms in Nigeria, the following hypotheses are

formulated:

Ho1: Board Competency does not have significant impact on Earnings Management of listed

foods and beverages firms in Nigeria.

Ho2: Frequency of Board Meetings does not have significant impact on Earnings Management of

listed foods and beverages firms in Nigeria.

Ho3: Gender Mix does not have significant impact on Earnings Management of listed foods and

beverages firms in Nigeria.

1.5 Scope of the Study

11
This study focused on studying the impact of Board Characteristics on Earnings Management of

foods and beverages firms listed on the Nigeria stock exchange (NSE) for the period 2007 to

2013. The choice of the period is because, it is within the period when codes of corporate

governance for public limited liabilities companies were introduced two times by the Securities

and Exchange Commission. The first code was introduced in 2003; while the second code was

introduced in 2011 as a replacement to the 2003 SEC code; and most of the features of these two

SEC codes are related to board‟s strengthening.

The study restricts itself to analysing quantitative data in determining the impact of Board

Characteristics on Earnings Management. Thus, secondary data obtained from respective

companies‟ financial statements and other companies‟ profiles were used.

1.6 Significance of the Study

The research result can help shareholders to know whether or not the variables of the study

(namely- Board Competency, Board Meetings and gender sensitivity) are effective in

constraining Earnings Management. This will enable them to make timely decision regarding

their investment and to take appropriate action to control the behaviour of the board.

The research findings may help shareholders of the foods, beverages, tobacco and breweries

firms in Nigeria to assess the level of compliance with the code of corporate of governance by

the management, as well as the effectiveness of board in monitoring managers‟ opportunistic

behaviours. The result may also provide information that is more valuable to Nigerian

accounting regulators in making recommendation for corporate governance practice.

The study would extend prior researches on board effectiveness by including Board Competency

of the directors as a determinant. It may be possible that directors achieve different performance
12
results under different board compositions. However, few studies have considered the Board

Competency. Therefore, the result might be useful for companies to design more effective board

composition.

The research result will be immensely beneficial to the regulatory authorities in Nigeria (like

Securities and Exchange Commission and Corporate Affairs Commission) in understanding

whether the variables studied in this work (Board Competency, frequency of Board Meetings and

Gender Mix) could be effective to control the level of earnings manipulation in foods, and

beverages firms in Nigeria. The result will be expected to provide a further policy on how to

constrain Earnings Management. The result can also be used to assess the effectiveness of Board

of Directors in Nigeria to provide a reference for future policy reform.

Due to dearth of literature or empirical studies on the impact of Board Competency, frequency of

Board Meetings and gender sensitivity on Earnings Management in Nigeria, the study will add to

available literature.

13
 CHAPTER TWO
LITERATURE REVIEW

2.1 Introduction

This chapter reviews literature relevant and related to the problems of our study. The review

covers conceptual issues related to Earnings Management, corporate governance mechanism,

board of director responsibility as specified in the Nigerian code of corporate governance and the

empirical studies on the nexus between various Board Characteristics (Board Competency,

frequency of Board Meetings and gender mix) and Earnings Management. The theoretical

framework relevant to the study has also been discussed. All these were done to identify a gap in

the literature to justify the need for the research.

2.2 Conceptualization

2.2.1 Concept of Board Competency

Competencies focus on what is expected of person in the work place and his or her ability to

transfer and apply knowledge and skill to their work. Previous researches show that directors

appear to require various clusters of competencies (Yusoff & Armstrong, 2012). The beginning

of the study of Board of Directors‟ competencies started in the United Kingdom upon the

establishment of Cadbury Report in 1992. Directors‟ competencies are seen to be of importance

to corporate governance (Yusoff & Armstrong, 2012). Hambrick & Manson (1984) reveal two

types of essential competencies required by the team of Board of Directors of firms- „functional

knowledge and firms‟ specific knowledge‟. Functional knowledge refers to knowledge in

14
accounting, finance, and legal marketing economics. While a firms‟ specific knowledge

incorporates the detail information about the firms and its operation. Therefore, directors who

had reasonable financial background are more effective in providing internal control system

mechanisms to control firm‟s performance. Cadbury (1992) states that non-executive member

that is competent is of special importance for effectiveness of the board. A director may acquire

the competencies through internal and external training and experience. Increase in the

proportion of non-executive directors does not guarantee the effectiveness of board monitoring.

But what guarantee is competency. Outside director must possess the necessary competency to

carry out their control and overseeing duties, among which knowledge of company specific

affairs is particularly essential (Chtorou, et al. 2001). Directors with financial expertise can

provide incremental control effects on Earnings Management (Man, 2012). Competencies

include functional and specific knowledge, skills and educational qualification that are necessary

to enable directors to perform their role of monitoring top management (Yusoff & Armstrong,

2012). Agrawal & Chada (2001) argue that manipulation of earnings tends to be low for firms

with directors who have knowledge in accounting and finance.

2.2.2 Concept of Board Meetings

Nowadays, there is a lot of criticism by many financial and academic publications on Board of

Directors for not attending Board Meetings regularly; and which by extension may lower their

ability to monitor management well (Modugu & Dabor 2013). The levels of board interaction

and activities have influence on Earnings Management. Boards that normally meet often are

more likely to solve the problems of the company effectively (Lipton & Lorsch, 1992). Lawler

and Conger (2001) suggest that number of times a board meet is an important resource in

improving the effectiveness of the board. Directors on board that meet frequently are more likely
15
to discharge their duties in accordance with shareholders interest because more time can be

devoted to monitoring issues such as Earnings Management, conflict of interest and monitoring

management. Diligent board enhances the level of oversight, resulting in improved financial

reporting quality. Jaiswal & Banerjee (2012) argue that more number of meetings would

facilitate more vigilant monitoring by the board in a company‟s affairs and thus would be

associated with better firm‟s performance and thus reduce Earnings Management. Lei (2004)

suggest that Board Meetings time is an important resource for improving the effectiveness of

board. Adams (2000) argues that the number of meetings of the Board of Directors would be a

good proxy for corporate governance to control the managers‟ behaviour. According to Jensen

(1993), frequency of Board Meetings is an important proxy for measuring the integrity and

effectiveness of corporate monitoring and discipline. This would constrain the level of earnings

manipulation. On the contrary, he argues that most of the Board Meetings are not effective since

the board is often forced to engage in high frequency activities to resolve corporate issues.

2.2.3 Concept of Gender Mix

Reasonable argument drawn from agency theory suggests that gender and ethnic diversity can

have either a positive, negative or neutral influence on a firm‟s performance (Carter, Simkims &

Simpson, 2010). Women participation in the corporate board has been increasing. In the past five

years, seven countries have passed legislation mandating female board representation and eight

have set non-mandatory target (Suisse, 2012). Pearce & Zahra (1991) argue that a representation

of diverse interest including the number of females and minority members is an important

characteristic of an effective board. Adams, Gray & Nowland (2010) find that the gender of

directors appears to be value-relevant and suggest that appointing female directors may help

resolve value-decreasing stakeholders‟ conflict. Diversity in agency due to diversity in groups

16
results in creating balance in the board and prevent from violence of members. It is high possible

that women members in the board would challenge about different subjects or specific

management decision by their questions and they would result into clear issues (Moradi, et al.

2013). Adams & Ferreia (2009) argue that female directors can better monitor managers‟

behaviour. Hence, they can improve the quality of firms earnings, as they have better

communication, used more informed decisions, and independent thinking (Tsui, et al. 2011).

Female directors can think more independently compared with male directors and also

effectively monitor CEO behaviour (Carter, et al. 2003). Furthermore, female directors are more

likely less tolerant than male directors to opportunistic activities under supervision (Man, 2012).

Tsui, et al. (2011) states that female directors can improve board governance which is more

likely to improve earnings quality. Therefore, female representation of board can actually reduce

the extent of Earnings Management because they can develop trust leadership, which requires

managers to share information, and they are risk averse for fraud and opportunistic Earnings

Management. On the contrary, some researchers suggest that women directors are only appointed

for symbolic rather than for substantive reasons.

Cornet & Warlard (2008) defined diversity as a set of personal, social, and organizational

characteristics that contribute to the development of identity and personality of individuals. At

the organizational level, it can be illustrated through an equal representation of men and women

in the top of the hierarchy as well as equal treatment that would guarantee social justice and

dismantle all form of discrimination (Campbell & Minguez-Vera, 2008). Women have been

gaining ground on corporate board, yet the effect of women on corporate performance is a matter

of some debates. A survey by catalyst (2009) shows that during the year 2008 and 2009, women

represented 15% of the board seats of fortune 500 companies; and 90% of these companies have
17
at least one woman in their corporate boards while 20% have more than 3 women. Studies using

data find that gender diversity on boards is associated with greater profitability. Adam & Ferreia

(2009) find that female directors can be better monitor of managers‟ behaviour through board

input such as attendance. Gulzar & Wang (2011) describe that the studies evidence the

emergence of an issue of board sex diversity in corporate governance literature started from the

last few years. Several studies have recently focused that the female member in the board can

affect the firms‟ performance. However, some studies suggest that firm performance has no

significant impact with board gender diversity. Carter, Simkins & Simpson (2003) argue that

women may improve decision making of the board. Fondas & Sassalos (2000) argue that

heterogeneous board is more efficient than homogenous board.

2.2.4 Concept of Earnings Management

Earnings Management assumes various terminologies: creative accounting, financial re-

engineering, and accounting magic. Schipper (1989) defines Earnings Management as disclosure

management in the sense of purposefully intervention in the external reporting process, with the

intent of obtaining some private gain. According to Healy & Wahlen (1999), the most important

reason why managers engage in the management of earnings is to enhance their compensation as

well as safeguarding their job.

This shows that managers are sometimes rewarded based on their contribution/ performance.

Therefore, there is likelihood that managers would engage in income-increasing Earnings

Management. This is possible due to flexibility in accounting standard. Apart from the reward

they could receive it may at the same time enable them safeguard their job from the eyes of the

18
law and the owner of the business. This is because owners have no in-depth knowledge of the

status of their investment.

Healy & Wahlen (1999) also defined Earnings Management as a situation that occur when

managers use their judgment in financial reporting and restructuring transaction to alter financial

reports to mislead the stakeholders about the actual and true economic performance of an

enterprises as well as to influence contractual outcome that depends on the reported earnings.

Earnings Management is an anticipatory step to keep away from defaulting in a loan agreement

reduce regulatory cost and increase regulatory benefit (Cornett, Marcus & Tehran, 2008).

Therefore, firms are expected to report positive earnings figure. Earnings Management do occur

for many reasons, some of which are to:

1. Manage earnings with the view of influencing stock market perception.

2. Increase their (managers) compensation and bonus

3. Avoid breach of loan agreement

4. Avoid regulatory intervention (Teoh, Welch & Wong, 1998)

5. Avoid negative earnings surprises (Matsumoto, 2002) with the management could engage

in Earnings Management to avoid reporting losses and earnings declines.

According to Teoh, et al. (1998), managers have several windows of opportunities to manipulate

earnings within the boundary of Generally Accepted Accounting Principle. They can choose an

accounting method to advance or delay the recognition of revenues and expenses, use

discretionary aspects of the application of the chosen accounting method, or adjust the timing of

asset acquisitions and dispositions to alter reported earnings.

19
Bunamin, et al. (2012) described that any attempt to manipulate earnings to mislead users‟

perception is regarded as unethical even though no accounting standards are violated. This is

because majority of respondent do not believe that earnings manipulation is ethical (Rafik,

2002).

According to Kassem (2012) many financial analysts believe that Earnings Management is

harmful. This is because it tends to conceal the actual result of the companies‟ earnings; and it

also shows that any attempt by the managers to use whatever means even if the means is

approved accounting standard, or otherwise, is considered to be unethical and it erodes the

quality of financial reporting.

According to Gulzar & Wang (2011), Earnings Management unlike fraud involves selection of

accounting procedure and estimates that conform to Generally Accepted Accounting Principle

(GAAP). It occurs within the bound of accepted accounting procedure. This kind of Earnings

Management that is consistent with Generally Accepted Accounting Principle (GAAP) is not

regarded as fraudulent financial reporting. Although, there is fear that management may cross

border from the true Earnings Management to fraud. However, once it can be determined that,

the management manipulated earnings with the intention of deceiving the perception of investors

is regarded as unethical and fraudulent act by many analysts.

Modugu & Dabor (2013) referred to Earnings Management as the use of accounting techniques

to produce financial reports that tend to present overly positive image of company‟s economic

performance. This indicates that companies employ income-smoothing techniques to smoothen

out fluctuation in earnings to meet analyst expectation. Based on their definition it indicates that

20
Earnings Management are normally adjusted upward to reflect persistent earnings or increased in

earnings.

Dechow & Skinners (2000) believe that accounting practitioners and regulators view Earnings

Management as a problem that needs an immediate control action. While some on the other hand

believe that Earnings Management practice by some firms benefits investors. Healy & Wahlen

(1999) argue that financial reporting can increase firms‟ value if economic earnings and firm

performance is reliable and available on time.

Opportunistic Earnings Management literature, originated with Healy (1985). The study

concluded that managers use accruals to strategically manipulate bonus income. For example,

managers can smooth earnings by deferring income through accruals when an earnings target for

a bonus plan cannot be reached or when bonus have already reached maximum level and can

accelerate income in other periods. Earnings Management may be defined as reasonable and

legal management decision making and reporting intended to achieve stable and predictable

financial result. Large numbers of companies are using Earnings Management to steady earnings

growth or to avoid reporting red ink. A common criticism of Earnings Management is that it

reduces transparency by obscuring the true earnings of the company.

2.3 Methods of Earnings Management

Amart, Blake & Dowds (1999) identified various methods of earnings management (Creative

Accounting)”

a. Accounting Choices: Accounting choices are made within the framework of Generally

Accepted Accounting Principle (GAAP). GAAP are set of rules, practices and convention
21
 that described what is acceptable financial reporting for external stakeholders. A firm is

allowed to choose between different methods of treating certain transaction within the

framework of Generally Accepted Accounting Principle (GAAP). Therefore, a company

can choose the accounting policies that give a preferred image. Managers are free to

choose an accounting estimates or method that would favour their reporting.

b. Artificial transaction can be entered into to manipulate monetary value of items in the

balance sheet and to move profits between accounting periods.

c. Genuine transaction can also be timed to give the desired impression in the account.

d. Certain entries in the account involve unavoidable degree of estimation, judgment,

and prediction without limit in the Generally Accepted Accounting Principles. The

problem with many accounting choice is that there is no clear posted limit beyond which

a choice is obviously illegal. Generally Accepted Accounting Principle does not tell

managers what specifically is normal and what is extreme. Product of warranty cost

estimation is an example of accounting decision many managers have to make.

Therefore, Earnings Management activities occur because of flexibility in making accounting or

operating choices and or/because managers are trying to convey private information to financial

statement users. Managers can opportunistically manipulate accounting reports by managing

accruals. According to Kaplan (1985), normal accruals arising from the ordinary course of

business may not reflect earnings manipulation. As such, any manipulation of earnings is likely

to manifest in abnormal accruals.

Most studies on Earnings Management focus on two types of Earnings Management: accrual

management and the manipulation of real economic activities. For accruals management, a firm

may decide to use provisions for warranty cost, inventory values, credit losses, and the timing of
22
an amount of unusual items. On the other hand, manipulation of real economic activities is

regarded as costly to affect firm‟s long-term interest. Schipper (1989) argues that it is very costly

to determine Earnings Management tactics. She further argues that even more visible Earnings

Management techniques like change in accounting policies and timing of capitalization are

difficult to interpret.

Recent studies show that top managers‟ compensation is linked to firm performance that is

associated to higher Earnings Management (Cornett, et al. 2008). Since mangers have inside

information, they have opportunities to manage net income to maximize their bonuses (Healy

1985) on the contractual motivation. Furthermore, managers may manage earnings to increase

net income at the expense of future earnings in order to secure their job (Defond & Park, 1997).

Also managers can manage earnings to avoid debt covenants. Firms are more likely to avoid

reporting losses by managing earnings; otherwise, they could violate covenants and face higher

cost. Also they may overstate earnings in order to meet analysts forecast (Man & Wong, 2013)

2.3.1 Earnings Management models

Accrual accounting is based on the notion that there is a difference between cost and

expenditures versus benefits and revenue. There are a number of subjective decisions involved in

the allocation of expenditures and revenue over time. In the Earnings Management literature, it is

often assumed that accruals are open to more discretion than cash flows. Several literatures

suggest various ways of detecting Earnings Management. Healy (1985) and Jones (1991) pointed

out that the most common way of detecting Earnings Management is the evaluation of accruals.

Earnings Management usually reflect in a firm‟s earnings. Accruals are examined in order to

determine whether they are discretionary or nondiscretionary accruals. The presence of abnormal

accruals indicates Earnings Management. Theorists have been attempting to deliver an accurate
23
and objective measurement for Earnings Management in order to do a better quantitative

research. The existing models of Earnings Management can be divided into three categories:

accruals model, specific accruals model and accruals distribution model (Liu, 2011). There are

many approaches in detecting Earnings Management but the Accrual-Based Models are the most

popular approaches. Analysis of Earnings Management often focuses on management‟s use of

discretionary accruals. In these accrual-based models, researchers estimate the discretionary

components of reported income. Accrual model is currently the mainstream measuring of

Earnings Management. This method points out that total accruals of a listed company is

composed of accruals by Earnings Management, namely discretionary accruals and non-

discretionary accruals. Therefore, from measuring non-discretionary accruals by constructing a

regression model, discretionary accruals is calculated as the difference between total accruals

and non- discretionary accruals, which measures the degree of Earnings Management.

There are several methods to measure abnormal accruals, which can be considered a proxy for

Earnings Management. The various models are discussed below:

1) The Healy model

Healy (1985) was the first person to attempt in the Earnings Management literature to estimate

Earnings Management by estimating deviations from normal levels of accruals. In his model, he

started with total working-capital accruals. He predicts that systematic Earnings Management

occurs in every period. He assumed that non-discretionary accruals follow random walk. For a

company in a stationary condition, the non-discretionary accrual in period t is equal to the non-

discretionary accrual in period t-1. As a result, the difference between the non-discretionary

accruals in period t and t-1 is the discretionary accrual that is related to Earnings Management.
24
Total accruals are defined by:

ACR = (ACAt - ACLt - ACasht+ ASTDt - Dept)/TAt-1

Where:

ACR= total working capital accruals

ACA = change in current assets

ACL= change in current liabilities

ACash = change in cash and cash equivalents

A STD = change in debt included in current liabilities

Dep = depreciation and amortization expenses

TA = total assets

2) The standard-Jones (1991) model

Jones (1991) believes that the variations of revenue would bring variations on operating capital,

causing a change in accruals, and the depreciations on fixed assets would decrease the accruals.

Because of this, Jones used variance of revenue (△ REV) and fixed asset (PPE), as independent

variables to predict the discretionary accruals. The Standard Jones (1991) Model uses a two-

stage approach to separate the normal and abnormal accruals. In the first stage, the parameters

are estimated by regressing the total observed accruals on the change in sales (ΔREV) and the

gross level of property, plant and equipment (PPE). The parameters are estimated for each

sample firm separately by using the longest available time-series data prior to the event year. In

the second stage, the discretionary component of the total accruals is determined by using the

parameters that have been estimated in the first stage of the model. In this stage, the total

expected accruals are determined by combining the parameters with ΔREV and PPE. The

abnormal or discretionary accruals (DAC) are determined by subtracting total expected accruals

from total observed accruals.

25
However, the model has its own shortcomings. One of the main disadvantages of the original

Jones (1991) is that time series estimation needs only firms with sufficient time-series data to be

included in Earnings Management research. For example, Jones (1991) selects firms with at least

10 time series data points for estimation. This requirement leads Earnings Management studies to

select only well established firms (usually large in size), while Earnings Management may also

exist in newly established and small firms. There are also problems related to serial correlated

residuals. That is, self-reversing property of accruals that may induce specification problems to

the Jones and modified Jones models (Peasnell, et al. 2000). The coefficient estimates on the

change in revenue and the property, plant and equipment (PPE) variables are unlikely to be

stationary over time; there are also confounding effects in the estimation periods, which are not

related to Earnings Management (Dechow, et al. 1995). Total accruals are defined as follows:

1 REVi ,t PPEi ,t
TACC / TAi ,t ˆ1 ˆ2 ˆ3 + et
TAi ,t 1 TAi ,t 1 TAi ,t 1

Where: TACC= total accruals

REV= change in revenue in year t

PPE= property, plant and equipment in year t

e= error term (Discretionary accruals)

TA= total asset

3) The modified-Jones (1991) model by Dechow, Sloan and Sweany (1995)

According to Dechow, Sloan and Sweeney (1995), the original Jones model is unable to capture

the impact of sales-based manipulation because accounts receivables should not be considered as

26
nondiscretionary accruals. Thus, they proposed a modification to the original Jones model known

as the Modified Jones model (1995). The modification is designed to eliminate a conjectured

tendency of the Jones Model to measure discretionary accruals with error when discretion is used

over revenues. Dechow, et al. (1995) give evidence that the Modified Jones Model currently

seems the most proficient at detecting Earnings Management. This is the most famous model to

detect Earnings Management nowadays. In Jones model and cross-section Jones model, the

assumption is that all the variances of revenue are non-discretionary. However, managers could

use credit sales to manage earnings. So the only adjustment of Modified-Jones Model is that it

deducts change in accounts receivable from change in sales revenue. The underlying assumption

is that, all the change in accounts receivable is caused by Earnings Management. This is based on

the following reasoning, for managers, it is much easier to manage earnings from accounts

receivable than from cash sale income. As pointed out by Peasnell, et al. (2000), this implies that

the model implicitly assumes that all changes in credit sales in the event period result from

Earnings Management. When comparing the original Jones model with the modified jones

model, Dechow, et al. (1995) find that the latter is significantly better in detecting sales-based

Earnings Management. However, both models show a significant proportion of Type I errors

when firm performance is extreme. Dechow, et al. (1995) explain that a Type I error occurs

when the null-hypothesis that earnings are not systematically managed is rejected, when it is

true.

1 REVi ,t RECi ,t PPEi ,t

TACC / TAi ,t ˆ1 ˆ2 ˆ3 et
TAi ,t 1 TAi ,t 1 TAi ,t 1

Where: TACC= total accruals

REV= change in revenue in year t

PPE= property, plant and equipment in year t

27
 REC= change in account receivable in year t

e= error term (Discretionary accruals)

TA= total asset

4) The cross-sectional version of the standard-Jones and modified- Jones model by

Defond and Jiambalvo (1994)
Jones uses time-series in the last model, but the data would incur bias. To avoid the bias, DeFond

and Jiambalvo (1994) introduced cross-section Jones model, which assumes the non-

discretionary accruals level in the same industry, are the same. First, the number of observations

per model is considerably higher under the cross-sectional version. This increases the precision

of the estimates. Second, by not imposing availability of time-series data, the cross-sectional

sample is less subject to survivorship bias and allows the researcher to include firms with short

histories. Third, misspecification of the coefficients due to non-stationary is not an issue for the

time series (Klein, 2006). The standard Jones and Modified Jones model require sufficiently long

time-series data in order to estimate the parameters in the first stage of the models. In practise,

this has proven to be quite restrictive as it introduces the chance of several problems occurring

with respect to survivorship bias, non-stationary coefficient estimates on ΔREV and PPE, and

serially-correlated residuals. The cross-sectional regression in each industry portfolio also

eliminates industry-specific effects when the accruals behaviour and the impact of the economy

vary across industries. However, the cross-sectional version of the models has several

weaknesses; for example, it is less probable that the model will capture: mean reversion in

accruals; dynamic accrual management strategies; and industry-wide Earnings Management

(Peasnell, et al. 2000).

28
The cross sectional versions of the two models are similar to the Jones (1991) and Modified

Jones (1995) models (see the two models above), except that the parameters of the models are

estimated by using cross section, not time-series data.

5) Industry Model by Dechow and Sloan 1991

Dechow and Sloan (1991) present industry model, which also eases the assumption that non-

discretionary accruals, are constants as Jones model does. However, the difference is that

industry model assumes that influence factors of non-discretionary accruals are the same

between different enterprises in the same industry. Industry model eliminates the difference of

non-discretionary accruals within different enterprises in the same industry, but if the changes of

non-discretionary accruals are rendered by enterprise specific environment factors, the industry

model cannot separate non-discretionary accruals and discretionary accruals accurately.

Industry model is as follows:

NDA= R1+R2 median (TA)

Where:

NDA= non-discretionary accruals

Median (TA) = the median of total accruals divided by total assets in year t-1

R1 and R2= industry-specific coefficients estimated from OLS model.

6) The margin model- by Peasenell, et al. (2000)

The margin model is applicable when cash flow performance is extreme. It uses a two-stage

approach to determine abnormal accruals. However, the explanatory variables in the first stage

regression are derived from a formal model linking sales, accruals and earnings. Depreciation is

not taken into account, as it is believed that it is not appropriate for systematic Earnings
29
Management. Consequently, working capital accruals (WCA) are used rather than total accruals.

WCA is regressed on total sales (REV) and total sales minus the change in trade debtors (REC).

Peasnell, et al. (2000) argues that the primary advantage of this model is improved economic

intuition, and therefore a more precise estimate of normal accruals. However, they also state that

the downside of the model is that Earnings Management may, itself contaminate the variable that

is being used, REV. Consequently, the margin model may be less powerful in detecting revenue-

based Earnings Management.

The model is presented below:

WCA/TA= b0 + b1REVt/TA + b2CRt/TA + Et

Where:

WCA= working capital accruals (difference between change in non-cash current assets

and the change in current liabilities excluding portion of long

term debt)

REV= revenue (total sales)

CR= cash received from customers (total sale minus the change in trade debtors)

TA= total asset

b0-b3= coefficient

7) Performance-match accruals model by Kathori, Leone and Wasley (2005)

Performance-matched accruals models are applicable when firm performance is extreme.

This performance matched discretionary accruals model is calculated by incorporating return on

assets (ROA) into the Modified Jones model. Dechow, et al. (1995) point out that the probability

of Type I errors occurring is significantly high when the (cross sectional version of the) Original

Jones and Modified Jones models are used in situations where firm performance is extreme.
30
They give two reasons for the occurrence of these Type I errors. Firstly, the nondiscretionary

accruals could be correlated to firm performance. Hence, the null hypothesis is falsely rejected

because of correlated measurement error in the proxy for discretionary accruals. Secondly,

Earnings Management could be caused by other factors that are correlated to firm performance.

In this case, Earnings Management is correctly detected; however, the cause of it remains

unknown. Hence, a factor could be chosen that does not cause Earnings Management, but is

correlated with firm performance; leading to a misspecification of the tests. Kothari, et al. (2005)

argues that their approach provides additional controls when firm performance is extreme.

Moreover, they argue that performance matching will remove performance motivated Earnings

Management, as both the treatment and match control firms experience similar performance.

Also, they state that performance-matched discretionary accruals represent abnormal accruals

instead of total accruals. Additionally, the results of their study show a preference for the

standard Jones model compared to the Modified Jones model under the performance matching

approach. They find that the Modified Jones model will spuriously find Earnings Management

under performance matching on the basis of ROA. Thus, bias is not eliminated from the

Modified Jones model in low sales growth samples, under performance matching on the basis of

ROA. Contrarily, bias is found to be eliminated from the Original Jones. The model is presented

below:

1 REVi ,t PPEi ,t
TACC / TAi ,t ˆ1 ˆ2 ˆ3 ei ROA
TAi ,t 1 TAi ,t 1 TAi ,t 1

Where: TACC= total accruals

REV= change in revenue in year t

PPE= property, plant and equipment in year t

REC= change in account receivable in year t

31
 ROA= Return on assets

e= error term (Discretionary accruals)

TA= total assets

2.4 Concept of Corporate Governance, Board of Directors and Agency Problems

Corporate governance has become a pressing issue since the 2000 introduction of Sarbanes-

Oxley Act in the United State which was issued in order to improve public confidence in

companies after financial scandals of Enron and Worldcom. After these crises, people began

thinking and questioning the effectiveness of corporate governance mechanism within an

organization. Subsequently, corporate governance and Board of Directors‟ responsibility have

been at the centre of the policy debate concerning governance reform.

Corporate Governance is a straightforward agency perspective because of the separation of

ownership and control. The agency problem is an essential element of the contractual view of the

firm developed by Coarse (1937) and Jensen & Meckling (1976). The essence of agency

problem is the separation of management and control. The agency problem in this context refers

to the difficulties a provider of finance faced in assuring that their funds are not expropriated or

wasted on investment. The creation of agency problem results from the separation of ownership

and control. As managers have more inside information than the financial providers; these

financial providers face agency cost to monitor managers‟ behaviour. The managers might

pursue their selfish interest to maximize their own wealth at the expense of other parties‟ wealth

and interest (Jensen, 1986).

32
Man and Wong (2013) classified corporate governance mechanisms into two types: external and

internal mechanisms. External mechanisms are determined by outside factors, intend to govern

firms in favour of the interests of stockholders; and it include items such as legal protection and

takeover rules. While internal mechanisms are determined by internal factors including insider

shareholdings, board structure and Board Characteristics, such as independent directors‟

background, audit committee and ownership structure. Therefore, corporate Governance has

succeeded in attaching a good deal of public interest because of its apparent importance for the

economic health of corporations and society in general. As a result, different people have come

up with different definitions that basically reflect the special interest in the field.

Corporate governance is the system by which business and firms are directed and controlled. The

cooperate government structure specifies the distribution of a rights and responsibilities among

different participant in the board, managers, shareholders and spell out the rules and procedures

for making decision on corporate affairs OECD (1999) and Cadbury (1992).

Cooperate governance is about promoting cooperated fairness, transparency and accountability

(Wolfenshohn, 1999). Nat (2013) describes that the prominence of Board of Directors in

cooperate governance is an evidence in model definition of corporate governance; which sees

corporate governance as the processes and structures by which the affairs of a companies are

directed and managed, for the purpose of improving the long-term shareholders value. This is by

enhancing corporate performance, accountability while taking into consideration the interest of

other stakeholders. This indicates that the primary responsibility of corporate governance is to

protect the interest of shareholders and this is vested on the shoulders of Board of Directors, as

they are responsible for managing and controlling the activities of a firm by influencing

management decision. The corporate governance exist to provide check and balance between
33
management and shareholders in order to reduce agency problem that many arise due to the

separation of ownership and control (Ibrahim, 2007).

Corporate Governance is viewed from two perspectives: Narrow view and broad view. The

narrow view (Anglo-Saxon) perceives corporate governance in term of issue regarding to

management control and shareholders protection based on the principal-agency relationship of

economic theory (Oyejide & Soyibo, 2001). Broad view (Franco-German) perceives corporate

governance as mechanism that protects the interest of stakeholders that is shareholders, creditors,

managers directors, government, society and any other interest party. That is why Raut (2010)

defined corporate governance as the process aiming to allocate recourses in order to maximize

the value of stakeholders-shareholders, investors, employees, customers, supplier, environment

and the society. From the foregoing definition we can perceive that corporate governance is not

only to protect the investors but all others parties that help the organization to move forward

directly or indirectly. Since their contribution towards achieving the organizational goal cannot

be overemphasize. However, proponent of narrow view argued that investors have a unique

relationship with the firm as they sunk their investment and potentially place at risk.

Corporate governance is a set of process, custom, policies, laws and institution affecting the way

companies are directed, managed and controlled. Corporate governance is a concept where

management supervision take place in the decision making process. In the decision making

process, corporate governance must be implemented; and one of its requirements is a firm‟s and

organisation management. Dabor & Modugu (2013) Defined corporate governance as a system

of rules, practice and process that is apply in controlling and directing the activities of a

34
company. Based on the definition, it shows that the affairs of a company are managed and

controlled systematically and procedurally not arbitrarily.

Corporate governance also refers to how Board of Directors manage the companies, taking into

consideration the impact of decision on shareholders, employees, customers, suppliers and

communities. The Board of Directors oversee the conduct of the business and supervise

management. Corporate status allows directors to delegate certain power to the officers such as

the CEO. The board delegates responsibility for the company‟s day-to-day affairs to the

executive. The board may also have audit and compensation‟s sub-committee of which the

members of the committee will be subset of the board and report back to the Board of Directors

on specific issues (Kawasaki, 2004). The Board of Directors‟ key purpose is to ensure

company‟s prosperity by collecting the companies‟ affair whilst meeting the appropriate interest

of the shareholders and other stakeholders. In addition to business and financial issues, Board of

Directors must deal with challenges and issues relating to corporate affairs (Kawasaki, 2004).

Based on previous literature, Earnings Management can be seen as a potential agency cost since

managers manipulate earnings to mislead shareholders and fulfil their own interests. Among the

set of corporate governance mechanisms, Board of Directors is often considered as the primary

internal control mechanism to monitor top management, and protect the shareholder interest.

Therefore the Board of Directors which is in charge of solving the agency conflicts between

mangers and shareholders should play a role in constraining the level of Earnings Management.

Furthermore, prior similar researches of financial reporting fraud suggest that effective board

monitoring helps to maintain the credibility of financial reports (Lei, 2003)

35
According to Raut, (2010) there are various model of corporate governance around the world;

and this differs according to the environment they operate: Anglo American Model- which tends

to emphasize on the management responsibility of maximizing the interest of shareholders.

While Coordinate or Multi-Stakeholders Model- which associated with continental Europe and it

emphasizes on maximizing all the stakeholders interest in general. That all those parties that

have relationship with the company in whatsoever must be satisfied accordingly that is the

interest of every user must be satisfied.

In summary, directors are responsible to ensure financial statement is prepared according to

approved accounting standard. Therefore, in order for the board to function effectively, certain

characteristics are suggested within the structure of corporate governance. The major

components of corporate governance include some benchmark of Board of Directors‟

characteristics. Therefore, main characteristic of Board of Directors refers to board composition,

board size, directors‟ ownership and duality status of the Chairman and CEO (Saleh, Iskandar &

Rahmat, 2005).

2.5 Nigerian Code of Corporate Governance and Board of Directors’ Characteristics

According to SEC (2011), the code is not intended as a rigid set of rules and that the

responsibility of ensuring compliance with or observance of the principles and provisions of the

code is primarily with the Board of Directors. The SEC has also clearly indicated that its

intention for replacing 2003 SEC code with the 2011 SEC code is consistent with international

best practice on corporate governance. Currently there are four corporate governance codes in

force in Nigeria: the 2011 SEC code and three others which are industry-specific. The three

others are: 1) the 2006 code of corporate governance for banks in Nigeria post-consolidation
36
issued by central bank of Nigeria and applicable to all banks operating in Nigeria. 2) The 2008

code of corporate governance for licensed pensions operation issued by the national pension

commission applicable to all pension fund administrators and pension fund custodians operating

in Nigeria. 3) The 2009 code of corporate governance issued by the national insurance

commission, and applicable to all insurance companies operating in Nigeria.

The following are the responsibilities of the board as prescribed in the SEC (2011) code: to

oversee the effective performance of the management in order to protect and enhance

shareholders value as well as meeting the companies‟ obligation to its employees and other

stakeholders; to define the company‟s strategic goal and ensure that both human and financial

resources are effectively deployed towards attaining those goals; it is responsibility of the board

to ensure that the company carries on its business in accordance with its articles and

memorandum of association and in conformity with the law of Nigeria; and the board shall

define a framework for the delegation of its authority or duties to management specifying matters

that may be delegated and those reserved for the board.

The board is an effective corporate governance mechanism. Shareholders elect members of the

board to act on their behalf, and the board in turn delegates power to top management while still

monitoring management performance and ratifying any decision that demonstrates a lack of good

faith for shareholders. Composing a Board of Directors is an important corporate governance

mechanism that can control managers‟ opportunistic behaviour and reduces Earnings

Management. Board composition includes determining the mix of non-executive (including

independent) and executive directors; designating audit, compensation, nominating and other

committees; determining the mix of qualifications and areas of expertise; and determining the
37
proportion of female directors on the board (Man & Wong, 2013). Bertrand and Mullainathan

(2000) also consider internal governance measurements, including the percentage of independent

directors on the board and the number of Board Meetings. Independent directors supervise the

managers in the firm and reduce the misappropriation of assets by the managers at the expense of

shareholders‟ interests. A higher percentage of the independent directors would more efficiently

control the managers, providing better corporate governance.

2.6 Review of Empirical Studies

The monitoring role of the Board of Directors is an important element of corporate governance.

Several studies that emerged have empirically examined the relationship between Board

Characteristics and Earnings Managements. Prior empirical researches suggest that some

characteristics of a board have an influence on the quality of financial reports (Chtourou, Bedard

& Courteau, 2001). Previous studies also recognize the importance of a board as an internal

control mechanism; and further suggest that an effective board reduces Earnings Management.

Hence, in the next sub-headings we are going to examine the empirical researches conducted by

some scholars on the relationship between various Board Characteristics and Earnings

Management. The variables of interest that includes Board Competency, Board Meetings and

Gender Mix are to the best of our knowledge captured the most crucial aspect of directors‟

attributes.

2.6.1 Board Competency and Earnings Management

38
In a survey carry out by Yusoff & Armstrong (2012), eight types of competencies were ranked to

be necessary and significant for Malaysian companies‟ directors. However, accounting and

finance knowledge was ranked to be the most essential competencies for directors.

This provides evidence that directors‟ knowledge of accounting and/or finance should have

significant effect Earnings Management. Directors with accounting and finance background may

have a better chance of understanding earnings manipulation when compare to directors who do

not have knowledge of accounting and finance (Johari, Saleh, Ja‟afar & Hassan 2008).

According to Xie, et al. (2003) directors who do not have appropriate competencies in

accounting and finance field may be able to monitor business and marketing process but may not

be able to understand Earnings Management practice. As such, there is an expectation that

directors who have accounting and finance knowledge are able to reduce the level of Earnings

Management.

Chtourou, et al. (2001) examines whether corporate governance practices have an impact on

Earnings Management. They grouped US firms into to two: firms with high and low levels of

discretionary accruals. They find that income decreasing Earnings Management is negatively

associated with the presence of at least one member of Board of Directors with financial

expertise. They also find that firms‟ outside directors that have experience as board members

tend to have less income increasing Earnings Management that is experience board seem to

reduce income decreasing Earnings Management. Therefore, the result finds evidence that

effective board constrain Earnings Management. It supports the view that experience board

member who has both experience on his company and other firms will provide better governance

competencies.
39
Bala & Gugong (2015) examines the influence of board characteristics and earnings management

of listed food and beverages firms in Nigeria. The study covers the period of six years 2009 to

2014. A multiple regression was employed to test the model of the study using Random Model.

The results from the analysis revealed an inverse relationship between board size, board

meetings and board financial expertise, and earnings management of listed food and beverages

firms in Nigeria, while and board composition and women directorship are positively

significantly related to earnings management of listed food and beverages firms in Nigeria.

Lin & Hwang (2010) employed meta-analytic techniques to the data from nearly 48 empirical

studies, and find a negative relationship between board financial expertise and earnings

management. In contrast Matawee (2013) show that a positive relationship between board

financial expertise and earnings management.

Johari, et al. (2008) analyse the influence of board independence, competency and ownership on

Earnings Management in Malaysia. By using dichotomous variable of 1 if a firm has at least one

director that has professional qualification in accounting and finance, and 0 otherwise as a

measure of Board Competency, while discretionary accruals as a measure of Earnings

Management for a sample of 234 firms listed on the Bursa stock exchange over the period 2002-

2003. The result reveals a non-significant relationship between knowledge as well as experience

of the board members and Earnings Management activities. But Agrawal & Chada (2001) find

that earnings manipulation in the US is low in firms with board members who have accounting

and finance knowledge.

40
Bunamin et al. (2012) examines the relationship between board diversity (size, independence,

competency, remuneration, and gender) and discretionary accruals of the top 100 companies in

Malaysian corporate governance index in 2008. By using number of directors with professional

qualification in accounting and finance as proxy for Board Competency, they find that Board

Competency, board independence, board remuneration, and leverage do not have significant

impact on discretionary accruals.

Yusof (2009) investigates to find out whether the Malaysian code of corporate governance

(MCCG) has achieved its goal in improving the quality of earnings among firms. The study uses

a sample of 190 KLCI components over a 4 years period from 1998 to 2003. The sample is

divided into two categories: pre-MCCG and post-MCCG periods. The study employs modified

Jones (1995) model to calculate discretionary accruals. In the post-MCCG, the result shows a

significant positive association between board competencies and discretionary accruals. He

suggests that longer tenure of service in board provides potential opportunities for them to

manage reported earnings.

2.6.2 Frequency of Board Meetings and Earnings Management

Jaiswal & Banerjee (2012) argue that more number of meetings would facilitate more vigilant

monitoring by the board in a company‟s affairs and thus would be associated with better firm‟s

performance and thus reduce Earnings Management.

41
Kantudu & Ishaq (2015) examines the impact of monitoring characteristics on financial reporting

quality of the Nigerian listed oil marketing firms. Financial reporting quality is represented with

the qualitative characteristics of financial statement. Data for the study were obtained from an

audited annual report and accounts of the sampled oil marketing companies for twelve years

covering 2000 to 2011. Multiple regression was used to analyzed the data. Board meeting has an

insignificant negative relationship with FRQ of the listed oil marketing companies in Nigeria in

both OLS and RE regressions. This implies that frequency of meeting increases the level of

earning manipulation which in turn decreases the quality of financial reporting. The finding is

contrary to our expectation.

Francis, Hasan & Wu (2012) use the current financial crisis as quasi-experiment to examine

whether and to what extent corporate boards affect the firm performance by using cumulative

stock returns over the crisis to measure firms‟ performance. The study finds that the frequency of

Board Meetings, directors‟ attendance behaviours, and director age (experience) affect firms‟

performance during the crisis.

Jaiswal & Banerjee (2009) examine the relationship between Earnings Management and

corporate governance characteristics in the Indian context. The result suggests that more number

of Board Meetings and higher attendance in these meetings resort to lesser Earnings

Management. DeMelo (2009) also examines the effect of the Dutch corporate governance code

on the practice of Earnings Management by publicly listed companies in Netherland by using a

sample of 33 firms over the period 1997 to 2006. He finds that frequency of Board Meetings

does not have significant relationship with Earnings Management.

42
Vafeas (1999) argues that the greater the meetings frequencies the more effective will be the

boards monitoring function. They reveal that companies that have fewer Board Meetings than

necessary, the value of firms will decrease. Xie, et al. (2003) finds evidence that Board of

Directors that meet more frequent are more likely to reduce the level of Earnings Management.

Previous studies evidenced that a board that meets more frequently is likely to monitor the board

activities in order to reduce the Earnings Management more effectively (Xie, et al. 2003; Vafeas

1999).

Gulzar & Wang (2011) examine the effect of corporate governance characteristics on Earnings

Management and find a significant positive association between Earnings Management and

different corporate governance characteristic such as CEO duality and board meeting. Vafeas

(1999) empirically investigates the relationship between board activities and the firm

performance. He uses frequency of Board Meetings as a proxy for board activities. He finds that

increase in Board Meetings lead to improved firm performance.

Sukeecheep, et al. (2013) investigates the influence of Board Characteristics on Earnings

Management behaviour in Thai listed companies. The sample of the study consists of 550 firms

for a period 2006 to 2010. The study finds no significant impact of Board Meetings on the

Earnings Management of top Thai listed firms. The expectation is that Board of Directors that

meet frequently will be more effective in monitoring the integrity of financial management and

therefore, more likely to constraints earning management.

Zhou & Chen (2004) investigate the impact of audit committee and Board Characteristics on

Earnings Management by using abnormal loan loss provision to measure Earnings Management
43
for a sample of over 300 listed firms in China over the period 2000-2002. The result reveals that

Board Meetings are negatively related to Earnings Management for low Earnings Management

banks. But for high Earnings Management group, Board Meetings, audit committee, board

independence and board size play an important role in constraining Earnings Management. They

suggest that board activity is an important dimension of board operation and effective board

should meet regularly to stay on top of accounting and control related matters. Hence, Board

Meetings among others are important mechanism of corporate governance.

Moradi, Salehi, Bighi & Najari (2012) investigate the relationship between Board Characteristics

and Earnings Management of listed companies on Tehran stock exchange during 2006 to 2009.

The result shows the frequency of Board Meetings does not have significant impact on Earnings

Management.

2.6.3 Gender Mix and Earnings Management

Adam & Ferreia (2009) find that female directors can be better monitor of managers‟ behaviour

through board input such as attendance. Gulzar & Wang (2011) describe that the studies

evidence the emergence of an issue of board sex diversity in corporate governance literature

started from the last few years. Several studies have recently focused that the female member in

the board can affect the firms‟ performance. However, some studies suggest that firm

performance has no significant impact with board gender diversity. Carter, Simkins & Simpson

(2003) argue that women may improve decision making of the board. Fondas & Sassalos (2000)

argue that heterogeneous board is more efficient than homogenous board.

44
Omoye & Eriki (2014) conducted a study to classify Nigerian quoted companies into high and

low earnings management levels and also to investigate how corporate governance mechanisms

relate to these categories of earnings management levels. A sample of 130 companies were

drawn from quoted companies on the Nigerian stock exchange over the period of 2005 to 2010

and to identify the unique firm‟s corporate governance characteristics. The study revealed that,

quoted companies in Nigeria prefer to use high earnings management practices; and that board

gender representation had a negative and significant influence on the probability of Nigerian

firms adopting absolute high earnings management.

Srinidhi et al. (2011) conduct a study on US firms over the years 2001-2007 and use accruals

quality (extended Dechow & Dichev model as proposed by McNichols (2002)) and target

beating as proxies for earnings quality. All the results show that female presence on the board is

associated with both higher accruals quality and less propensity to manage earnings to beat

benchmarks. They conclude that firms with women on the board have higher earnings quality.

Gavious et al. (2012) use a sample consisting of 60 Israeli high-technology firms listed in the

USA between 2002 and 2009 to investigate whether gender diversity on the board affects

earnings quality. They analyse four earnings quality metrics of which two are ex-post measures

(abnormal accruals and non-operating accruals) and two are ex-ante measures (presence of a Big

Four auditor and financial leverage). Their results suggest that female presence on the board is

associated with better earnings quality.

Oscar & Daniel (2013) studied Swedish listed companies within the period 1999-2007. In order

to measure earnings quality, the extended Dechow & Dichev model (McNichols, 2002), and

45
performance-matched abnormal accruals were used. While boardroom gender diversity is

measured as the proportion of female directors on the board found no association between

boardroom gender diversity and earnings quality. Ye et al. (2010) investigates if female presence

on the board promotes higher earnings quality in Chinese firms, and collect data of firms listed

on the Chinese A-share market from 2001-2006. They follow Dechow & Schrand (2004) and

consider earnings to be of high quality when they are persistent, indicative of future cash flows,

and related to current stock performance. Results from the study suggest that female board

members have no significant influence at all on earnings quality.

Bunamin et al. (201) examines the impact of board diversity and discretionary accrual of the top

100 companies in Malaysian Corporate Governance index. They find that Malaysian Corporate

Governance index companies manage earnings despite their best compliance on corporate

governance. They also find that women on board have positive relationship with discretionary

accruals that is increases in the proportion of women in the board would lead to increase in

Earnings Management. Also O‟Reilly and Main (2012) examine the effect of women outside

directors on firm performance and CEO compensation. Using a sample of over 2000 firms for

the period 2001 to 2005, the result shows no evidence that adding women outsiders to the board

enhances corporate performance.

Hili & Affes (2012) empirically study the impact of gender diversity of the Board of Directors on

earnings quality and on Earnings Management by using a sample of 70 French firms. The result

reveals that the enhancement of earnings persistence could not attribute to gender diversity that is

gender diversity does not have significant impact whether positive or negative on Earnings

Management. Also Carter, D‟Souza, Simkims & Simpson (2010) investigate the relationship
46
between the number of women directors and the number of ethnic minority directors on the

board and financial performance. They find no significant relationship between the gender and

ethnic diversity of the board and financial performance for a sample of major US Corporation.

Emilia & Sami (2010) examine the association between gender, firms‟ executives and Earnings

Management for a sample of 500 Finland firms. Panel regression of discretionary accruals is

used. The result provides a considerable evidence to suggest that firms with female directors are

associated with income-decreasing accruals. This implies that females are following more

conservative Earnings Management strategies. The result of the empirical analysis indicates that

the gender of the firms‟ executive may affect the quality of financial reporting. Also, Rose

(2007) find that no significant association between female board representation and firm

performance.‟

Gulzar & Wang (2011) investigate the efficiency of corporate governance characteristics in

reducing Earnings Management among listed firms of Shanghai and Shenzhen stock exchange in

China. The result shows that presence of female director in the board help to reduce the level of

Earnings Management.

Tsui, et al. (2011) investigate the relationship between female participation and corporate boards

and earnings quality drawn from a sample of US listed firms from 2007 to 2011. They employ

two measures of earnings quality: discretionary accruals and propensity of firms to beat earnings

benchmark by small amount. The result of the study shows that the presence of female directors

in monitoring position on audit and corporate governance committees makes more transparent

reporting and earnings quality. It also reveals a positive relationship between female

participation in corporate boards and earnings quality. Furthermore, after controlling for
47
endogeneity, and other firms and industry characteristics, the result still shows a higher earnings

quality in firms with female board participation. Also Niskanen, et al. (2009) reveals that

earnings quality is positively associated with gender diversity in corporate management.

Springer (2008) examines whether and how the participation of women in the firm‟s board of

directors and senior management do not enhances firm‟s financial reporting or ability of

checkmating opportunistic behaviour of management. The study finds that women are often

appointed to leadership positions under problematic organizational circumstances associated

with greater risk of failure and criticism, however, having more women on corporate board and

top management does not seem to generate significant excess return and cannot restrained

managers‟ opportunistic behaviour.

Ghazaleh & Garkaz (2015) examine the relation between the presence of women on the boards

of directors of listed companies in Tehran Stock Exchange using and earnings management with

discretionary accruals index. The required information were extracted from 90 accepted firms in

Tehran Stock Exchange using Cochran sampling method for 7-years period (2006-2012). The

findings of the study indicate that the presence of female directors in board is significantly and

negatively associated with earnings management. It was made clear that firms with women on

their boards, they have less use discretionary accruals for earning management.

Krishnan & Parsons (2005) examine the relationship between the proportions of women in the

senior management rank and earnings quality by using a sample 770 firms-year observations for

a period 1996 to 2000. The result shows that earnings quality is higher for firms with higher

48
gender diversity in senior management than for firms with low gender diversity in senior

management.

Sun, Liu & Lan (2011) investigate whether female directors represented on the independent audit

committees affect the ability of the committees in constraining Earnings Management. Using a

sample of 525 firm-year observations for a period 2003 to 2005, the study did not find any

significant association between the proportions of female directors on audit committees and

extent of Earnings Management.

Wei & Xie (2010) examine the relationship between CFO gender and Earnings Management by

using a sample of publicly traded firms from 1999 to 2006. The study finds that male CFOs

engage more Earnings Management than female CFOs. It also reveals that female CFOs are

more risk averse in making financial reporting and operational decisions than male CFOs are.

2.6.4 Literature Gap

Despite the fact that there are many empirical studies that investigated the issue of Board of

Directors‟ attributes/corporate governance and Earnings Management, majority focused on

determining the statistical impact of board size, board independence, CEO duality, managerial

ownership and audit committee on Earnings Management. However, Board Competency,

frequency of Board Meetings and Gender Mix have not yet been studied extensively.

The empirical studies reviewed especially on the effect of Board Competency on Earnings

Management cast some doubt on the statistical result. Despite there are many arguments

supporting the view that directors‟ knowledge of accounting and finance helps in constraining

Earnings Management, yet most of the empirical researches find insignificant impact of Board

Competency on Earnings Management. Some even find significant positive association. This
49
could be as a result of the fact that these studies (like that of Chtorou et al. (2001) and Johari et

al. (2008)) use dichotomous variables to represent Board Competency, by assigning 1 if at least

one member of Board of Directors has qualification in accounting and finance and 0 if otherwise.

This will not capture the proportion of Board of Directors with accounting and finance

knowledge in each firm. Furthermore, Bunamin et al. (2012) use number of Board of Directors

with accounting and finance knowledge as a proxy for Board Competency, forgotten that board

size varies from one firm to another. Some companies may relatively have larger board size than

another. Therefore, the measurement could not capture the relative proportion of Board of

Directors with accounting and finance knowledge. Hence, we measured Board Competency in

term of „ratio of Board of Directors with qualification in accounting and finance to total board

size.

Furthermore, on the issue of gender sensitivity and Earnings Management, there is clear

evidence that some group of companies or sectors are more gender sensitive. Most of these

empirical researches studied the entire listed companies in their respective countries. This may

not capture the real impact of women in the boardroom. However, taking a particular domain

that almost have similar business and gender sensitivity can provide a real result. That is also

another reason for restricting our research to foods and beverages sector to see the influence of

gender sensitivity on Earnings Management in that particular sector.

Most of the empirical studies reviewed that considered these three variables if not those of Bala

& Gugong (2015), Kantudu & Ishaq (2015) and Omoye and Erike (2014) among others are

foreign based studies. As such, further study is required to ascertain how these variables can

50
influence earnings management in Nigeria with reference to listed foods and beverages firms in

Nigeria.

2.7 Theoretical Framework

The theory underpinning this study is agency theory. This is the theoretical framework used by

most researchers for clear understanding of the impact of board characteristic and firm value

(cater, Simkins & Simpson, 2003). Agency theory elucidates the existence of the incentive for

management to use Earnings Management (Salah, 2010). Jensen & Meckling (1976) define an

agency relationship as a contract under which one or more person (the principals) engage another

person (the agent) to perform some services on their behalf which involves delegating some

decision making authority to the agent. According to Brennan (1995) the agency problem usually

arises if an agent fails to act in the best interest of the shareholders (the principal). This may

occur when managers try to achieve their own interest at the detriment of the shareholders; just

to maximize their reward or meet some earnings target or debt covenant. All this is possible

because of separation of ownership and control and information asymmetry (since managers

have more information than the real owners of a company). Management may manage earnings

to hide the true financial position and relevant information of a business organization that

investors ought to have known. Agency theory elucidates the possibility for managers to manage

earnings; managers may produce a bias financial report without the opportunity of others to see

through it. Because of the opportunistic behaviour of an agent, corporation put in place a

mechanism to align the interest of the principal and agent through the establishment of Board of

Directors (Bunamin et al. 2012). Therefore, their action as an agent of shareholders may provide

better monitoring of management that in turn leads to transparent and quality reporting.

Shareholders employ team of Board of Directors to oversee the effective performance of the
51
management in order to protect and enhance their value as well as meeting the companies‟

obligation to its employees and other stakeholders.

In most agency relationships, the principal and the agent will incur monitoring and bonding

costs. In a corporate entity, to oversee the management operation and constraint the management

opportunistic behaviour, the shareholders invest in information and monitoring system, including

employing Board of Directors, audit committee, and auditors (Hashim & Devi 2008).

However, impact of corporate governance mechanism on Earnings Management should be

placed in the context of agency problem arising from the separation of ownership and control.

According to Fama and Jensen (1983), creating a board that is effective in monitoring

management action depends on the composition of individuals serving on the board. Thus,

certain characteristics of Board of Directors are suggested in the literature to be effective in

constraining the level of Earnings Management.

Agency theory suggests that a more diverse board may provide better monitoring of

management. Furthermore, board of diverse gender may better avoid practice of Earnings

Management, hence, provides shareholders with more reliable financial reporting (Gallego,

Garcia & Rodriguez, 2010). Agency theory also supports the idea that corporate boards that meet

more frequently have increased capacity to effectively advise, monitor and discipline

management, and thereby enhancing corporate performance (Ntim & Osei, 2011)

52
 CHAPTER THREE
STATEMENT OF METHODOLOGY

3.1 Introduction

This chapter discusses the methodology adopted for the study. It explains the research design,

population of the study, sample size and sampling techniques adopted in the study. It describes

the method of data collection, analysis and interpretation as well as variables measurement,

research instrument and technique of statistical analysis of data.

3.2 Research Design

The study adopted correlational research design by employing both descriptive and inferential

statistics using regression analysis. A correlational study tries to measure the degree of

relationship between one or more variables for making predictions about relationship. In order to

use the correlational research design, the variables of the study must be related; each variable of

the study must be expressed in numerical form, that is, it must be quantifiable; and the research

must follow a quantitative approach (Water, 2014; Ramscar, 2002).

In addition to correlational research design, the study also adopts ‘‘ex-post facto or causal-

comparative design’’. Data are collected after the event or phenomenon under investigation has

taken place, which is why it is called ex-post facto. That is, the causes are studied after they have

presumably exerted their effect on another variable before the research.

Descriptive statistics is used to present and summarise data for easy understanding and

interpretation; while inferential statistics is also used for data analysis and interpretation in such a

way that conclusion about the population would be reached, based on data received on the

sample (Vincent, Olaegbe & Sobona, 2008).

53
The choice of correlation design is because the study aimed at finding the relationship between

Board Characteristics and Earnings Management; the study also attempted to follow quantitative

approach to find the effect of Board Characteristics on Earnings Management (all the variables

of the study have been assigned numerical values in the process of establishing the relationship).

Ex-post facto design was also adopted because the study investigated the effect of independent

variables- Board Characteristics on the dependent variable- Earnings Management after the event

under investigation has taken place. Ex post facto design is a quasi-experimental study

examining how independent variables present prior to the study affect a dependent variable.

3.3 Population of the Study

The population of the study comprised all the listed foods and beverages firms on the Nigerian

Stock Exchange as at 2013. Therefore, the population comprises all the twenty-one (21) listed

foods and beverages firms in Nigeria. The total numbers of firms are derived from the Nigerian

Stock Exchange (NSE) fact book as at 2013. The classification based on the fact book is as

follows:

Beverages-brewers/distiller 7

Beverages-non-alcoholic 1

Food products 11

Food products diversified 2

Total 21

(See appendix 1 for the list of companies)

Data for the study was obtained from the annual financial statements of the listed companies,

NSE factbook and other companies‟ profiles.

54
3.4 Sample size and Sampling Technique

The sample size of the study is nine (9) foods and beverage firms listed on the Nigeria stock

exchange. Effort was made to see that adequate sample size was drawn, considering the size of

the population (21) as well as the degree of precision desired. The sample size covered 43% of

the population. This is arrived at after using 2 stages criteria:

The first stage is that a firm must meet the criterion of being listed on the NSE within 2007-2013

and should not have been delisted within the period. As such Honey well flour mills plc and

Nigeria Bottling Company were deselected, and we were left with nineteen (19). The second

stage is that a firm must have been publishing it directors‟ profile as well as having information

on variables of the study. Therefore, the sample size that covers the span of this study and

satisfies the criteria of having information on all the variables at the time of conducting this

research were nine (9) listed foods and beverages firms. Thus, some companies were dropped

because we could not get access to Board of Directors profiles (issue related to Board

Competency), while some companies do not have complete financial statements within the

period to get the information related to accounting figures required for computation of

discretionary accruals, ROA and firm size.

The period for the study is 2007 to 2013. The period reflects the time when corporate governance

is in effect in the Nigerian capital market. This is because the regulatory innovation began in

2003 and subsequently re-strengthens in 2011.

3.5 Method and Sources of Data Collection

55
The study used secondary source of data collection. The data was collected from the annual

reports and account of the sampled companies, Nigerian stock exchange factbook and other

relevant sources for a period of seven (7) years (2007 to 2013). The firms are public limited

companies listed on the Nigerian Stock Exchange. By virtue of being public limited companies

and as a requirement of being listed, annual financial report has to be made available to the

Nigerian Stock Exchange.

3.6 Techniques of Data Analysis

Multiple regression was used to analyse the relationship between Board Characteristics and

Earnings Management. The data for the study is panel in nature (that is cross-sectional time-

series data). Panel data lead to errors that are clustered and possibly correlated overtime. This

makes the use of OLS regression problematic.

The regression was run in a panel manner; as such, various options of panel data regression were

run, like, random effect GLS regression, fixed effect (within) regression and GLS regression. But

the most robust of all is fixed effect (within) regression (as suggested by Hausman specification

test). Therefore, by extension, the result is reported using Fixed Effects-Least Square Dummy

variables (LSDV). The Fixed Effect- LSDV provides a good way to understand fixed effect (see

Model Specification). Stata 10 was used as a tool of analysis because it allows for carrying out

certain statistical tests, like heteroskedasticity test, VIF test, Hausman test, and estimation of

residuals.

3.7 Model Specification

56
The essence of the model is to study the impact of Board Characteristics on Earnings

Management of listed foods and beverages firms in Nigeria. From the literature reviewed, the

study found that Board Competency, frequency of Board Meetings and Gender Mix are

important attributes of Board of Directors; and play a significant role in monitoring managers‟

opportunistic earnings management behaviour. Therefore, statistical analysis for this study has

its root in agency theory. Hence, the model is stated as follows:

DACit = b0 + b1BCit + b2FBMit + b3GMit +b4 ROAit + b5FSIZEit + Et

Moreover, the model of the study is further modelled with Fixed Effect- Least Square Dummy

variables (LSDV). This is because „fixed effect within regression‟ was suggested by Hausman

Specification Test. The Fixed Effect- LSDV provides a good way to understand fixed effect. The

effect of independent variables is mediating by the different across sampled firms. Hence, by

adding the dummy for each firm, the model estimated the pure effect of Board Characteristics on

Earnings Management that is, by controlling for the unobserved heterogeneity. Each dummy is

absorbing the effects particular to each firm. As such, dummy variables is created for each firm

(except one), and included in the model. Since we have nine (9) firms each of whom was

measured at 7 points in time (7 years), then we included eight (8) dummy variables in the model.

The statistical inferences is drawn using this model. So the model for the Fixed Effect- LSDV

becomes:

DACi=b0+b1BCi+b2FBMi+b3GMi+b4ROAi+b5FSIZEi+b6D1+b7D2+b8D3+b9D4+b10D5+b11D6+b

12D7+b13D8+Ei

Where:

DAC = Discretionary accruals

b0 = intercept (constant)
57
b1 – b6 = parameters (slope coefficient)

BC= Board Competency

FBM = Frequency of Board Meetings

GM = Gender Mix

ROA = Return on Asset

FSIZE = Firm Size

D1-D8= dummy variables for each firm (except one)

3.8 Variables Measurement- Independent and Dependent Variables

Dependent Variable Measurement

Earnings Management (Discretionary The absolute values of discretionary accruals

Accruals) (residual obtained from modified Jones model (by
Dechow et al. (1995))

Independent Variables
Board Competency Proportion Board of Directors who have
qualification in accounting or finance to total
Board of Directors.

Frequency of Board Meetings Natural logarithms of number of Board Meetings

held in a year.

Gender Mix Proportion of female Board of Directors to total

Board of Directors.

Control Variables
Firm Size Natural logarithms of total asset.

Return on Asset (ROA) EBIT/Total Asset

Source: compiled by the author (2014)

Dependent Variable: Discretionary accruals measurement

58
This study used discretionary accruals to measure Earnings Management. Modified Jones (1995)

model was used to extract the discretionary accrual, that is, the residual values obtained after

regressing in the modified Jones Model. According to Dechow, et al. (1995), the original Jones

model is unable to capture the impact of sale-based manipulation because account receivables

should not be considered as non-discretionary accruals. Thus, they proposed a modification to

the original Jones model known as modified Jones Model (1995). The study adopts modified

Jones (1995) model because it is widely tested and accepted by many scholars and it is the best

model to estimate discretionary accruals with minimal error (Bello and Yero, 2011).

Furthermore, both Gulzar and Wang (2011) and Johari et al. (2008) used this model to estimate

the extent of Earnings Management. The model is presented below:

1 REVi ,t RECi ,t PPEi ,t

TACC / TAi ,t ˆ1 ˆ2 ˆ3 +et
TAi ,t 1 TAi ,t 1 TAi ,t 1

Where = TACC = Total accrual (Discretionary + Non-discretionary) for firm ‟i‟ in year t

REV = Change in operating revenues for firm i in year t

REC = Change in account receivables for firm i in year t

PPE = Gross property, plant and Equipment for firm i in year t

t and t-i = time subscript

i = firm subscript

e= error term (residual value) - discretionary accruals.

Therefore, to separate non-discretionary accruals component from discretionary accruals

component in the total accruals, the stages involve:

59
The total accruals have to be computed first. Therefore, in our study total accruals are computed

by the differences between net incomes and operating cash flows (that is net income for firm i in

year t – net operating cash flow for firm i in year t). Then insert the value of TACC (Total

accruals) into the modified Jones model‟s equation. PPE are also included to control for the

portion of total accruals related to non-discretionary depreciation expenses.

After regressing for the above equation, the residual value (eit) obtained, represent

discretionary component of the total accrual. Therefore, the absolute values of the residuals

represent Earnings Management

3.9 Control variables

In this study, firm size and return on assets (ROA) were used as control variables to control for

the effects of Board Characteristics on Earnings Management. These are variables often used by

many researchers when determining the impact of Board Characteristics on Earnings

Management.

ROA was used as a control variable by Klein (2002); Bartov, Givoly and Hayn (2001); Gulzar

and Wang (2011) and Johari, et al. (2008). ROA gives an idea as to how efficient management is

at using its assets to generate earnings. Dechow et al. (1995) suggest that firm performance

supposed to have a positive association with discretionary accruals. Without controlling for such,

discretionary accruals may reflect changes in the sample firm performance. Johari, et al. (2008)

finds that ROA has a significant positive relationship with discretionary accruals. Also Moradi,

et al. (2012) finds that ROA has a significant positive impact on Earnings Management. Riahi &

Arab (2011) also found that profitability has a positive relationship with Earnings management.

60
Firm size is used in most Earnings Management studies to control for many factors such as

political cost and economies of scale. Earlier studies have found that firm size has negative

impact on Earnings Management (Dechow, et al. 1995; Defond & Jiambalvo 1994). Therefore,

discretionary accruals are expected to have a negative relationship with firm size. According to

Abed, Al-Attar and Swaidan (2012) smaller companies are subject to less control from authority

and therefore, engage in Earnings Management activities but some argue that Earnings

Management activities increases as the size of a company increases. Soliman and Ragab (2013)

have empirically found that firm size has a significant positive relationship with Earnings

Management. They claim that large firms face greater scrutiny from investors, and thus more

likely to manage earnings to satisfy their forecast. Bolouri, et al. also (2011) documents a

positive relationship between firm size and Earnings Management. Lee (2013) in his study, „the

influence of corporate governance on earnings quality‟, finds that firm size does not have

significant impact on Earnings Management. In addition, Johari, et al. (2008) does not find any

significant relationship between firm size and Earnings Management. On the contrary, Musa, et

al. (2013) documents a significant negative impact of firm size on Earnings Management.

3.10 Justification of the Methodology

The choice of correlation design is because the study aimed at finding the degree of association

between Board Characteristics and Earnings Management. While ex-post facto design was also

adopted because the study investigated the effect of independent variables- Board Characteristics

on the dependent variable- Earnings Management after the event under investigation has taken

place (the study relied upon historical data). The period of the study chosen reflects the time

when code of corporate governance is in effect in the Nigerian capital market. The study
61
followed quantitative approach to determine the impact of Board Characteristics on Earnings

Management. Therefore, in line with our objectives and hypothesis formulated we made use of

multiple regressions to determine the impact of independent variables on the dependent variable,

because it is the most suitable techniques for determining the extent of impact of independent

variables on dependent variable. Stata Statistical Package was used because it allows

ascertaining the impact of the independent variables on the dependent variable as well as testing

for robustness such as heteroskedasticity test, fixed and random effect test, multicollinearity test

etc. The variables for the study are Board Competency, frequency of Board Meetings Gender

Mix and discretionary accruals. In addition, Return on Asset and firm size are introduced in the

model as control variables. We used stata application package to generate residuals from

modified Jones (1991) model for each year of the study. The choice of modified Jones (1991)

model is because it is widely tested and accepted by many scholars and it is the best model to

estimate discretionary accruals with minimal error (Bello & Yero, 2011).

62
 CHAPTER FOUR
DATA PRESENTATION, ANALYSIS AND INTERPRETATION

4.1 Introduction

This chapter covers data presentation, analysis, interpretation, and discussion of the results of the

study. Hence, results from the descriptive statistics, the correlation matrix, the robustness test,

and the regression results are presented and discussed. The chapter also discusses the

implications of the research findings.

4.2 Descriptive statistics

Table 4.1 presents descriptive statistics of the variables of the study. The mean, standard

deviation, minimum, and maximum have been used to describe the data.

Table 4.1: descriptive statistics

Variables Obs Mean Std.Dev. Minimum Maximum Skewness Kurtosis
DAC 63 0.0574 0.0527 0.0018 0.2537 1.5676 5.6724
BC 63 0.3179 0.1130 0.0767 0.5000 -0.3739 2.2455
FBM 63 1.5249 0.1999 0.6931 2.0794 -0.7048 6.9657
GM 63 0.0711 0.0712 0.0000 0.2500 0.7184 2.9856
ROA 63 0.1788 0.1826 -0.3076 0.6117 -0.7060 3.9142
FSIZE 63 17.5408 1.153 14.8458 19.3514 -0.6710 2.3642
Source: Extracted from STATA output 10

63
The absolute value of discretionary accruals of the listed foods and beverages firms in Nigeria

(based on the modified Jones model) has a minimum value of 0.0018, (which is much closer to

zero) and a maximum of 0.2537. This shows that the minimum percentage of discretionary

accruals from the total accruals of the sampled firms is 0.18% (which is less than 1%). While

during the period, some firms manage earnings to the extent of 25% of the total accruals. The

magnitude of absolute value of discretionary accruals in the sample firms has a mean of 0.0574

with standard deviation of 0.0527. This indicates that the deviation between companies is quite

very small. The discretionary accrual values are above zero, which suggests the existence of

Earnings Management in the listed foods and beverages firms in Nigeria.

The Board Competency is measured by the proportion of Board of Directors who have

qualification in accounting and/or finance to total Board of Directors. From the table, the Board

Competency has a mean of 0.3179, a minimum of 0.07669 and a maximum of 0.5000. This

indicates that in each board of the sampled companies there is at least one board member with

qualification in accounting and/or finance. It also appears that 8% to 50% of the directors during

the period 2007-2013 were competent. This shows that there are directors who are competent

during the period of the study.

The number of meetings held by the Board of Directors of listed foods and beverages firms

during the period has a mean of 4.6824 (1.5249) with standard deviation of 0.9127 (0.1999) and

the range is from the minimum of 2 (0.6931) to a maximum of 8 (2.0794). Furthermore, the SEC

code states that, for Board of Directors to effectively perform its oversight function and monitor

64
managers‟ performance, they should at least meet 4 times in a financial year. From the foregoing,

we can see that some companies did not conform to the provision of the code (for seating two

times). But some companies have gone beyond the minimum requirement (for sitting up to 8

times). However, on average, the sampled companies have conformed to the provision of the

code for having a mean of 4.6825.

Gender Mix refers to the number of female directors divided by the total number of directors on

the board. From the table, we can see that the proportion of women on the boards range from a

minimum of 0.0000 to a maximum of 0.2500 and a mean of 0.0712. This shows that the women

representation on the board of the sampled companies is very poor. This is because the average

representation is only 6.4%. The minimum of 0.000 proves that some companies did not include

women in their team of Board of Directors during this period of study; and for those that

included women, the maximum representation is only 25%. The standard deviation of 0.0712

indicates the deviation between companies is low. This shows that representation of women on

the board is poor across the sampled companies. However, representations of women tend to be

low even in the developed economies (Catalyst, 1999).

The return on asset (ROA) is used as control variable. The ROA has a mean of 0.1718 with a

standard deviation of 0.1826, a minimum of -0.3076 and a maximum of 0.6117. This shows that

the profitability between the companies varies greatly as some companies reported negative

ROA of -30.76% while some company (ies) reported positive ROA as high as 61.17%. The

variance in ROA of the listed foods and beverages firms in Nigeria may sometimes differ across

65
the companies due to size in their total assets. The standard deviation of 0.1826 proves that ROA

between companies varies; as such, there is an expectation that the magnitude of Earnings

Management will differ across the companies. This is because ROA is scaled by total size (firm

size); and firm size is believed to have an effect on the magnitude of Earnings Management. On

average, the sampled firms reported positive ROA of 17.18%.

The second control variable is firm size proxies by natural logarithms of total assets. It has a

mean of 17.5408 with standard deviation of 1.153 and the range is from a minimum of 14.8485

to a maximum of 19.3514.

4.3 Correlation matrix

The correlation matrix is used to determine the degree of association between independent

variables and dependent variable. It is also used to identify whether there is relationship among

the independent variables themselves, to be able to detect if multicollinearity problem exists.

Table 4.2: Correlation matrix

Variables DAC BC FBM GM ROA FSIZE
DAC 1.0000
BC 0.4898 1.0000
FBM -0.2620 -0.0278 1.0000
GM -0.1793 0.2463 -0.2202 1.0000
ROA 0.2937 -0.1001 0.1608 0.2184 1.0000
FSIZE 0.0616 -0.2907 0.2037 -0.1334 0.4097 1.0000
Source: Extracted from STATA output 10

From the table, the relationship between Board Competency and Earnings Management of listed

foods and beverages firms in Nigeria is positive and significant at 1% level. This can be

observed from the correlation coefficient of 0.4898 with a significant value of 0.0000. The

positive association indicates that Earnings Management may increase with the increase in the
66
proportion of Board of Directors who have qualification in accounting or finance. Despite the

fact that it is found in the literature that Board Competency reduces Earnings Management, yet to

our surprise, the statistical result reveals a positive association. This may be possible, because

many people consider Board of Directors of public companies as passive entities that are

controlled by management (Lei, 2003). However, the conclusion cannot be made now because it

is only an association. To access the real impact, we have to conduct regression analysis.

Moreover, the frequency of Board Meetings has a significant but weak negative association with

Earnings Management of listed foods and beverages firms in Nigeria. This is because; the

correlation coefficient is 0.2620 significant at 5% level. The direction of the relationship

suggests that more meetings by Board of Directors may have a reducing effect on the level of

Earnings Management of listed foods and beverages firms in Nigeria.

Furthermore, the degree of association between Gender Mix and Earnings Management is -

0.1793. This shows that there is a negative association between the two. Although the association

is weak, but the correlation coefficient indicates that, there is an inverse relation between the

two. This statistical association suggests that adding woman in a board may reduce the incidence

of Earnings Management. The correlation coefficient will not serve as a basis for generalization

on the impact of Board Characteristics on Earnings Management as it only gives a degree of

association between the dependent variable and the independent variables.

67
Additionally, the ROA introduced as a control variable has a significant positive relationship

with Earnings Management of listed foods and beverages firms. This can be confirmed from the

correlation coefficient of 0.2937. While a positive association is observed between firm size and

Earnings Management, but statistically insignificant. This can be seen from the coefficient value

of 0.0616 with a probability value of 0.6315.

The relationships between independent variables themselves suggest to be minimal because, only

ROA and Board Competency have significant relationship with firm size. In order to access the

presence of multicollinearity, the study further conducted multicollinearity test, using Variance

Inflation Factor (VIF) and its reciprocal (1/VIF). The benchmark for VIF is that at 5%

collinearity is suspected, at over 10% collinearity is assumed to be present. The result suggests

absent of multicollinearity. This can be confirmed from the statistical result that shows all the

VIF and 1/VIF are above 1 and less than 1 respectively, while the mean value of VIF is 1.25. All

these suggest absence of multicollinearity

4.4 Robustness test

In order to improve the validity of the statistical inference, and to avoid making wrong

inferences, some robustness tests were conducted. For the purpose of the study, panel regression

was used; as such, various options of panel regression were run. These include OLS regression,

GLS regression, random effect GLS regression, fixed effect (within) regression and Hausman

specification test (see table 4.5). However, the most robust of all is fixed effect (within)

regression.

68
Table 4.3: Multicollinearity test
Variable VIF 1/VIF

fsize 1.36 0.732710

roa 1.36 0.734847
gm 1.26 0.795121
bc 1.16 0.862730
fbm 1.12 0.890670

Mean VIF 1.25

Table 4.4: Heteroskedasticity Test

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of dac

chi2(1) = 11.48
Prob > chi2 = 0.0007

Table 4.5: Hausman test

Test: Ho: difference in coefficients not systematic

chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 23.59
Prob>chi2 = 0.0003
(V_b-V_B is not positive definite)

The OLS regression result reveals that all the independent variables except firm size are

statistically significant with F-statistics of 10.50 and probability value of 0.000. There is an

absence of multicollinearity (see table 4.3). However, the Breusch-Pagan test for

heteroskedasticity reveals the chi-square value of 11.48, significant at 1% level (see table 4.4).

This indicates the presence of heteroskedasticity in the data. It is for this reason that OLS

technique has a high chance of resulting in inefficient statistical result and by extension leading

to incorrect inferences, which may lead to type 1 or a type 2 error.

69
Given that the data has heteroskedasticity problem, the study focuses on two techniques used to

analyse panel data- random effect GLS regression and fixed effect regression. When using fixed

effects model, it is assumed that something within the individual may affect or bias the predictor

or outcome variables and there is need to control for this. Fixed effect removes the effect of those

time-invariant characteristics so that the net effect of the independent variables on the dependent

variable can be assessed. While random effects assume that, the entity‟s error term is not

correlated with the predictors that allow time-invariant variables to play a role as explanatory

variables. A quantity being random means that it fluctuates over unit in some population; and

which particular unit is being observed depends on chance.

To decide between fixed or random effects the study run hausman test. The hausman used to test

whether the unique error term is correlated with the regressors. Fixed effects is the alternative

hypothesis which states that the entity‟s error term is correlated with regressor; while random

effects is the null hypothesis which states that the entity‟s error term is not correlated with the

regressor. Therefore, the results from random effect and fixed effect reveal that F-statistics are

significant in all scenarios. But hausman specification test considered fixed effect (within)

regression as the appropriate estimator of parameters on the bases that fixed effect correlated

with the variables as the hausman test is statistically significant at 1%. This can be confirmed

from the chi-square value of 23.59 and a probability value of 0.0003 (see table 4.5 above).

Therefore, hausman test indicated that the fixed effect (within) regression is the preferred model

that should be used. However, by extension the study run for Fixed Effect- LSDV regression

model. This model is arrived at by introducing dummy variables (as explained under Model

70
Specification). The dummy variables have the ability to control firms‟ individual intercept or

unobserved heterogeneity, that is, each dummy is absorbing the effects particular to each firm.

Hence, result will be interpreted using Fixed Effect- Least Square Dummy Variables Model. By

comparison, the fixed effect using dummies and fixed effect (within group) regression yield

same result; except that F-statistics slightly increase by 0.7 with fixed effect using dummies (see

appendix III and IV for comparison.

4.5 Regression results

This subsection deals with inferential analysis of data. The result is in relation to the impact of

Board Characteristics on Earnings Management of listed foods and beverages firms in Nigeria. It

contains individual impact of the independent variables (Board Competency, frequency of Board

Meetings and Gender Mix) on the dependent variable (Earnings Management). Hence the

regression results are presented below (table 4.3)

Table 4.6: Summary of Regression Results

Fixed Effect- LSDV
Variables Coefficient t-statistics
Constant -0.1089 -0.56 (0.580)
BC 0.0806 0.79 (0.4350)
FBM -0.0953 -4.03 (0.0000)
GM -0.2243 -2.70 (0.009)
Control Variables
ROA 0.2845 4.57 (0.000)
FSIZE 0.0212 1.45 (0.1530)
R-square 71%
Adj-R square 63%
F-Statistic 9.01 (0.0000)
Source: Extracted from STATA output 10

71
4.6 Test of Hypotheses

The regression results reveals that Board Competency has no significant impact on Earnings

Management of listed foods and beverages firms in Nigeria. This can be confirmed by the t-value

of 0.79 with significant value of 0.435. This shows that Board of Directors‟ qualification in

accounting or finance does not have any significant impact on the Earnings Management of the

sampled companies. This is contrary to the findings in literature which show that members of

Board of directors who have accounting and/or finance‟ background are better able to reduce

Earnings Management. The implication of this finding is that director‟s qualification in

accounting and finance does not have significant effect on Earnings management. The finding

serves as a base for failing to reject the null hypothesis earlier formulated, which states that

Board Competency does not have significant impact on Earnings Management of listed foods

and beverages firms in Nigeria. The finding is in tandem with those of Johari et al. (2008),

Buamin et al. (2012) and Xie et al. (2003) who also found insignificant impact of Board

Competency on Earnings Management. It is however, contradicts the finding of Agrawal and

Chada (2001) who found that earnings manipulation in the US is low in firms with board

members who have accounting and finance knowledge.

Moreover, the inferential statistics show that frequency of Board Meetings has a t-value of -4.03

with a significant value of 0.0000 (that is significant at 1% level). This signifies that frequency of

Board Meetings is negatively influencing the Earnings Management of listed foods and

beverages in Nigeria. This implies that the higher the Board Meetings the lower the discretionary

accruals. The beta coefficient of -0.0953 indicates that discretionary accruals reduce by 9.53%

72
with an increase in one more board meeting. This is in line with the argument put by Vafeas

(1999), Jaiswal, and Bernajee (2012) that the greater the meeting frequencies the more effective

will be the boards‟ monitoring function and subsequently constrain Earnings Management. The

implication of this finding is that foods and beverages firms‟ compliance to Board Meetings will

enhance earnings quality by reducing the level of discretionary accruals. The result provides

justification for the rejection of the null hypothesis, which states that frequency of Board

Meetings has no significant effect on Earnings Management of listed foods and beverages firms.

The finding is in line with those of Jaiswal and Banerjee (2012), Xie. et al. (2003), Zhou and

Chen (2004) and Vafeas (1999) who also found a negative impact of Board Meetings on

Earnings Management. But the finding contradicts those of Sukeecheep et al. (2013), Moradi et

al. (2012) and DeMelo (2009) who did not find evidence of a significant relation between

frequency of Board Meetings and Earnings Management.

The regression results further reveal that Gender Mix has negative and significant impact on

Earnings Management of listed foods and beverages firms at 1% significant level. This can also

be observed from the statistical t-value of -2.7000 with probability value of 0.009. This shows

that increase in the proportion of women directors on the board significantly reduces the

Earnings Management of listed foods and beverages firms in Nigeria. By implication therefore,

even though the percentage of women on the board of listed foods and beverages firms is very

low, an increase in female representation on the board will enhance earnings quality through

constraining the level of discretionary accruals. The beta coefficient of -0.2242 indicates that

discretionary accruals will reduce by 22.42% with an increase of at least one-woman director.

This finding confirms the assertion made by Tsui et al. (2011) that female participation in a

73
board can improve earnings quality because they have better communication; use more informed

decision and independent thinking and have a developed trust leadership. The finding serves as a

base for the rejection of the null hypothesis, which states that Gender Mix has no significant

impact on Earnings Management of listed foods and beverages firms in Nigeria. Interestingly,

this finding is similar to the findings of Gulzar and Wang (2011), Tsui et al. (2011) and

Niskanem et al. (2009). The result is contrary to those found by Hili and Affes (2012), Sun et al.

(2011) and O‟Reilly and Man (2012) who did not find evidence of a significant impact of gender

diversity on Earnings Management.

The control variable-ROA has a significant positive impact on Earnings Management of listed

foods and beverages firms in Nigeria. The t-value of 4.57 and the probability value of 0.0000

suggest the positive impact. This means that managers of the firms are more incline to manage

earnings when performance is high. Dechow et al. (1995) and Kazniks (1999) suggest that firm

performance is expected to have a positive association with discretionary accruals. The finding is

in line with those of Johari et al. (2008) and Moradi et al. (2012) who found that ROA has a

significant positive relationship with discretionary accruals. Ria and Arab (2011) find a

significant positive association between ROA and Earnings Management; and believe that their

result is consistent with the prediction of several researches who affirm that Earnings

Management is operated in the case where an extreme performance is achieved.

The second control variable is firm size. The impact of firm size on Earnings Management of

listed foods and beverages firms in Nigeria is positive, but statistically insignificant because the

74
t-value is 1.4500 with a probability value of 0.153. Even though, it is not significant the positive

impact suggests that large firms manage earnings than small firms. Nevertheless, the finding of

the study is in tandem with those of Lee (2013) and Johari (2008) who did not find evidence of a

significant relation between firm size and Earnings Management.

The F-statistic as shown in the table measures the overall impact of independent variables (Board

Competency, frequency of Board Meetings, Gender Mix, ROA and firm size) on the dependent

variable-discretionary accruals. The result shows that F-statistic is 9.01 significant at 1% level.

This implies that the board characteristic variables investigated in the study have a significant

combined impact on Earnings Management of listed foods and beverage firms in Nigeria during

the period under review (2007-2013). It also suggests the fitness of the model. The R-square-

coefficient of determination is 0.705. This explained the individual variation of the dependent

variable because of change in the independent variables. Therefore, the Board Characteristics,

proxies by Board Competency, frequency of Board Meetings and Gender Mix (in addition to

ROA and firm size) have combined predictive power of 71% in controlling Earnings

Management of listed foods and beverages firms in Nigeria. While a negligible 29% is attributed

to stochastic error variances. The adjusted R-square provides an adjustment to the R-square such

that independent variables that have correlation with dependent variable increases adjusted R-

square. Therefore, the R-square adjusted for the number of variables that have strong

correlation in the model occupies 63% of the Earnings Management.

75
For the fact, the R-square is 71%, and the F-statistics is 9.01 (significant at 1% level), it indicates

the model is well fitted. Moreover, based on this we can conclude that Board Characteristics

have significant impact on Earnings Management of listed foods and beverages firms in Nigeria.

CHAPTER FIVE
SUMMARY, CONCLUSION AND RECOMMENDATIONS

5.1 Summary

As a response to financial scandals and some corporate failures in Nigeria linked to managerial

incentive to manage earnings, SEC to improve the monitoring function of the Board of Directors

introduced the code of corporate governance for 2003 and 2011. Therefore, this study

investigated the effect of some Board Characteristics namely, Board Competency, frequency of

Board Meetings and Gender Mix on Earnings Management (in the context of agency relation) of

listed foods and beverages firm in Nigeria from 2007 to 2013. The study considered the

discretionary accruals of the firms to represent the extent of Earnings Management. The

estimation of discretionary accruals was based on the modified Jones (1991) model. The study

hypothesized that Board Competency, frequency of Board Meetings and Gender Mix have no

significant impact on Earnings Management of listed foods and beverages firm in Nigeria. ROA

and firm size were included in the model of the study as firm specific factors that may have

influence on Earnings Management.

76
The sample size of the study is nine (9) out of twenty-one (21) firms. Both correlational and ex-

post factors research design were used. This is because the study tried to find the relationship and

effect of Board Characteristics on Earnings Management using historical data. The data collected

was purely from secondary sources; extracted from the companies‟ annual reports and other

corporate websites. A multiple regression technique was employed to assess the impact of Board

Characteristics on Earnings Management. The regression was run in a panel manner; various

options of panel data regressions were run but the most robust of all is fixed effect (within)

regression. This by extension the study run for Fixed Effect- LSDV regression model. The

dummy variables have the ability to control firms‟ individual intercept or unobserved

heterogeneity that is each dummy is absorbing the effects particular to each firm.

The study found that Board Characteristics have significant impact on Earnings Management of

listed foods and beverages firms in Nigeria. Individual results showed that Board Competency

has no significant impact on Earnings Management of listed foods and beverages firm in Nigeria.

The impact of frequency of Board Meetings on Earnings Management of listed foods and

beverages firm in Nigeria was found to be negative and statistically significant. Gender Mix was

also found to have significant negative impact on Earnings Management of listed foods and

beverages firm in Nigeria. Lastly, the control variable, ROA was found to have significant

positive impact on Earnings Management, while firm size was statistically insignificant in

influencing the level of Earnings Management of listed foods and beverages firm in Nigeria.

5.2 Conclusions

Based on the findings of the research, the study concludes as follows:

77
i. The empirical findings in the study provide evidence that Board Characteristics have

significant impact on Earnings Management of listed foods and beverages firm in

Nigeria. The empirical findings revealed that two individual characteristics of Board of

Directors (frequency of Board Meetings and Gender Mix) have a significant negative

effect on Earnings Management of listed foods and beverages firm in Nigeria, while

Board Competency does not have significant influence on Earnings Management.

ii. The study confirms the importance of corporate board meetings in enhancing the

monitoring function of Board of Directors. The significant negative effect of frequency of

Board Meetings on Earnings Management was an indication that more frequency of

Board Meetings reduces Earnings Management of listed foods and beverages firms in

Nigeria. This is possible when board members meet frequently to discharge their duties in

accordance with shareholders interest. Therefore, firms with board of directors‟meetings

that are more frequent are associated with low Earnings Management.

iii. Moreover, the significant negative impact of Gender Mix on Earnings Management of

listed foods and beverages firms in Nigeria was an indication that Earnings Management

reduces with an increase of women in the team of Board of Directors. This is possible

because women do normally challenge management decision by their question; and they

always develop trust leadership than their male counterpart. Hence, more women

participation in the affairs of the board reduces the incidence of Earnings Management.

78
 iv. Furthermore, the empirical finding reveals that Board Competency does not have

significant impact on Earnings Management. This means that directors‟ knowledge in the

field of accounting or finance do not make any difference in the Earnings Management

practice of listed foods and beverages firms in Nigeria. Thus, Board Competency does

not guarantee the quality of earnings. The result did not support the notion that, financial

experts, because of their knowledge in accounting and/or finance are at better chance to

detect any form of earnings manipulation, they may use their expertise to promote

earnings management.

v. In general, these findings suggest that frequency of Board Meetings and Gender Mix

undertake less accrual management. Although, not all the variables of the study opposed

to the stated hypotheses, but the study has achieved its objectives by answering the

research questions.

5.3 Recommendations

In line with the findings and conclusions of this study, the following recommendations are

offered:

i. The study recommends that SEC should encourage adherence to at least the minimum

requirement (four times in a financial year) by making it mandatory to hold meetings at

least four times in a year. The SEC should encourage more frequent meetings by Board

of Directors because, the empirical evidence indicates that Board Meetings are associated

with low Earnings Management.

79
ii. SEC should henceforth consider gender composition when designing or amending the

provision of the code. It should provide a means of enforcing women participation on

corporate boards. There is need for her to specify the proportion of women a company

should at least maintain as directors and to make it mandatory for companies to abide by

it.

iii. The authority should provide a means through which more meetings will be encouraging

by enforcing public firms to be following the provision of the code accordingly. Hence,

appropriate measures need to be taken by government to ensure companies abide by the

regulation. This will be possible if government clearly shows its concern by compelling

companies to abide by the provision of the code.

iv. Government in collaboration with Corporate Affairs Commission should come up with a

policy that will force public companies to provide seat for women in their boards, give

them responsibilities in area of finance, and control related matters. As the study

established the importance of women in reducing discretionary accruals, women

participation would enhance firm performance and constrain earnings management, since

they normally develop trust leadership.

v. Lastly, the flexibility of the SEC code contributes failure and bankruptcy of some firms

in Nigeria as only 40% of the listed firms recognize the code put in place. Therefore,

there is need for the regulatory authority to make it rigid rule; and failure to abide by it

should attract appropriate sanctions.

80
5.4 Limitations of the study

Like most studies, this research work is subject to a number of limitations. The limitations

include:

i. The limitation of the study is the issue of survivorship bias related to the period of study;

as only foods and beverages firms being listed on the NSE between 2007 and 2013 were

included. Hence, firms with short histories (not listed within the period) were excluded.

The study should have used all the twenty-one firms, but dearth of data and survivorship

bias hindered the use of all the firms.

ii. There are various models for detecting the extent of earnings management, but this study

restricts itself to the use of Dechow et al. (1995) model. This is because, it is the model

used by most researchers (including Gulzar and Wang (2011) and Johari et al. (2008))

and was proved to produce minimum error.

iii. In the literature, Board Competency has been defined in two ways: Board Competency in

term of directors‟ knowledge in the field of accounting and/or finance and Board

Competency in term of directors‟ years of experience on a board. This study examined

the influence of Board Competency in term of directors‟ knowledge of accounting and/or

finance on earnings management.

5.5 Suggestions for Further Research

81
i. This study examined the effect of Board Competency, frequency of Board Meetings and

Gender Mix on Earnings Management of listed foods and beverages firms for the period

2007 to 2013. The period of the study can be extended to 2015; there are other Board of

directors attributes that were not captured in this study. All these need to be revisited to

capture some changes that might have occurred.

ii. The same Board Characteristic‟ variables used in this study can be examined in another

domain/sector to see whether or not the research findings would be similar.

iii. Further research should examine the effect of Board Competency (in term of directors‟

years of experience on a board) on Earnings Management.

iv. Further research is needed on the impact of independent director‟s competency on

earning management.

82
References

Abdul Ra‟uf, F. A., Johari, N. H., Bunamin, S. B., & Rahman, N. R. (2012). The impact

company and board characteristics and earnings management. Global review of

Accounting and Finance, 3 (2), 114-127.

Abed, S., Al-Attar, A., & Suwaidan, M. (2012). Corporate governance and earnings

management: Jordanian evidence, International Business Research, 5(1).

Adams, R., & Ferreira, D. (2009). Women in the boardroom and their impact on governance and

performance. Journal of Financial Economics, 94, 291-309

Adams, R. B. (2000). What do boards do? Evidence from committee meeting and director

compensation. Working Paper, New York.

Adams, R., Gray, S., & Nowland, J. (2010). Is there a business case for female appointments?

Working Paper, City University of Hong Kong.

83
Agrawal, A., & Chada, R. K. (2001). Do some outside directors play a political role? Journal of

Law and Economics, 44, 179–199.

Agrawal, A., & Chadha, S. (2005). Corporate governance and accounting scandals. Journal of

Law and Economics, 48(2), 371-406.

Akenbor, C. O., & Ibanichuka, E. A. L. (2012). Creative accounting practices in Nigeria banks.

An International Multidisciplinary Journal, Ethiopia, 6(3).

Bala, H., & Gugong, B. K. (2015). Board characteristics and earnings management of listed food

and beverages firms in Nigeria. European journal of Accounting, Auditing and Financial

Research, 3 (8), 25-41

Bartov, E., Givoly, D., & Hayn, C. (2001). The rewards to meeting or beating earnings.

Bello, A., & Yero, J. I. (2011). Debt reliance and earnings management in the listed

conglomerates. Nigerian Journal of Accounting Research, Ahmadu Bello University, Zaria,

Nigeria, 7(2).

Bertrand, M., & Mullainathan. S. (2000). Enjoying the quiet life? Managerial behaviour

following anti-takeover legislation. MIT, Working Paper.

Brennan, M. J. (1995). Corporate finance over past 25 years. Financial Management, 24(2) 9-22.

84
Cadbury, S. A. (1992). The code of best practices: Report of the committee on the financial

Aspects of corporate governance, Gee and co. Ltd.

Campbell, K., & Minguez-Vera, A. (2008). Gender diversity in the boardroom and firm financial

performance. Journal of Business Ethics, 83, 435–451.

Carter, D. A., Simkins, B. J., & Simpson, W. G. (2003). Corporate governance, board diversity

and firm value. Financial Review, 38(1), 33-53.

Carter, D. A., D‟Souza, F., Simkins, F., & Simpson, W. G. (2010). The gender and ethnic

diversity of US boards and board committees and firms financial performance: Corporate

governance. An International Review, 18(5).

Catalyst (2009). Catalyst census: fortune 500 women board directors.

Http://www.catalyst.org/publication/

CBN (2006). Code of corporate governance for banks in Nigeria post consolidation.

Christopher, W. (2009). Lecture 16 on generalized least square.

Chtourou, S., Bedard, J., & Courteau, L. (2001). Corporate governance and earnings

management. Working Paper.

Coase, R. (1937). The nature of the firm, Economics, 4, 386-405.

85
Conger, J. A., Finegold, D., & Lawler, E. (1998). Appraising boardroom performance. Harvard

Business Review, 76, 136-164.

Cornett, M., Marcus, A., & Tehranian, H. (2008). Corporate governance and pay performance:

The impact on earnings management. Journal of Financial Economics, 87.

Credit Suisse research institute (2012). Gender diversity and corporate performance.

Dabor, E. L., & Okoeguale, I. P. (2013). An evaluation of the implication of earnings

determinants in the banking industry: The case of Nigeria. African Journal of Social

Sciences, 3(3).

Dchow, P., & Dicheve, I. (20002). The quality of accruals and earnings: The role of accrual

estimation error, The Accounting Review, 77, 35-59.

Dechow, P., & Skinner, D. (2000). Earnings management: Reconciling the views of accounting

academics, practitioners and Regulators. Accounting Horizons, 14, 235-250.

Dechow, P., Sloan, R. G., & Sweeny, A. P. (1995). Detecting earnings management. The

Accounting Review, 70 (2), 193.225.

Defond, M., & Park, C. (1997). Smoothing income in anticipation of future earnings. Journal of

Accounting and Economics, 23, 115-139.

86
Defond, M. L., & Jiambalvo, J. (1994). Debt covenant violation and manipulation of accruals.

Demelo, G. L. (2009). The corporate governance and its implication for earnings management: a

study of the Dutch situation. Erasmus University Rotterdam.

Ebrahim, A. (2007). Earnings management and board activity: Additional evidence. Review of

Accounting and Finance, 6(1), 42-58.

Emilia, P., & Sami, V. (2010). Female executives and earnings management. Department of

Accounting and Finance, university of Vaasa. www.emiraldinsight.com.

Fama, E. F., & Jensen, M. (1983). Separation of ownership and control. Journal of Law and

Economics, 26, 301-325.

Financial Accounting Standard Board (1990). Technical Report, FASB Review Committee.

Fondas, N., & Sassalos, S. (2000). A different voice in the boardroom: how the presence of

women directors affects board influence over managerial. Global Focus, 12, 13-22.

Francis, B., Hassan, I., & Wu, Q. (2012). Do corporate boards affect firm performance? New

evidence from the financial crisis.

Gallego, I., Garcia, I. M., & Rodriguez, L. (2010). The influence of gender diversity on corporate

performance. Account Review, 13(1), 55.

87
Ganiyu, Y. O., & Abiodun, B. Y. (2012). Impact of corporate governance on capital structure

decision of Nigerian firms. Research Journal in Organizational Psychology and

Educational Studies, 1(2), 121-128.

Ghazaleh, E., & Garkaz, M. (2015). The association between presence of female members on the

board of directors and earning management with discretionary accruals index on the

listed companies in Tehran stock exchange. Journal of Current Research in Science, 3 (1), 39-

43

Gavious I, Segev E, Yosef R. (2012). Female directors and earnings management in high-

technology firms. 2012: Pacific Accounting Review, Vol. 24 No. 1, PP: 4-32

Gul, F. A., B. Srinidhi and J. Tsui: 2007, „Do female directors enhance corporate board

monitoring? Some evidence from earnings quality‟, Working paper, The Hong Kong

Polytechnic University

Gulzar, M. A., & Wang, Z. (2011). Corporate governance characteristics and earnings

managements: Empirical evidence from Chinese listed firms. International Journal

of Accounting and Financial Reporting, 1(1), 133-151.

Hambrick, D., & Mason, P. (1984). Upper echelons: the organization as a reflection of its top

managers. Academy of Management Review, 9, 193-206.

88
Hashim, H. A., Salleh, Z., & Ariff, A. M. (2013). The underlying motives for Earnings

Management. International Journal of Trade, Economics and Finance, 4 (5).

Hashin, H. A., & Devi S. S. (2008). Board independence, CEO duality and accrual management:

Malaysian evidence. Asian Journal of Business and Accounting, 1(1), 27-46.

Healy, P. (1985). The effect of bonus schemes of accounting decision. Journal of Accounting

and economic, 7.

Healy, P. M., & Wahlen, J. (1999). A review of the earnings management literature and its

implication for standard setting. Accounting Horizons, 13(4), 365-383.

Hendry, K., & Kiel, G. C. (2004). The role of the board in firm strategy: Integrating agency and

organisational control perspectives. Corporate Governance. An International Review,

12(4), 500–520.

Heninger, W. G. (2000). The Association between auditor litigator and abnormal accruals. State

university of New York, Buffalo.

Hermalin, B. E., & Wesberch, M. S. (2012). Information disclosure and corporate governance.

Journal of finance, LXVII (1).

Hili, W. & Affes, H. (2012). Corporate board gender diversity and earnings persistence: the case

of French listed firms. Global Journal of Management and Business Research.

89
Ibrahim, S. S. A., & Mohamad, H. S. (2012). The effectiveness of corporate governance on

constraining earnings management: literature review and proposed framework.

International Journal of Global business, 5 (1), 17-35

Jaiswal, M., & Banerjee, A. (2012). A study on the state of corporate governance in India:

exploring the relation between Earnings Management and corporate governance

characteristics in the Indian context.

Jensen, M. C., & Meckling, W. H. (1979). Theory of the firm, managerial behaviour agency, cost

and ownership structure. Journal of Financial Economics, 3, 305-360.

Jensen, M. C. (1993). The modern industrial revolution, exit, and the failure of internal control

systems. Journal of Finance, 48, 831–880.

Johari, N. H., Sale, N. M., Jaffar, R., & Hassan, M. S. (2008). The influence of board

independence, competency and ownership on earnings management in Malaysia.

Journal of Economics and Management, 2(2), 281-306

Kantudu, A. S., & Ishaq, A. S. (2015). Board characteristics, independent audit committee and

financial reporting quality of oil marketing firms: evidence from Nigeria. Journal of

Finance, Accounting and Management, 6 (2), 34-50

90
Kaplan, R. S. (1985). Comments on Paul Healy: Evidence on the effect of bonus schemes on

accounting procedure and accrual decisions. Journal of Accounting and Economics,

7, 109-113.

Kassem, R. (2012). Earnings management and financial reporting fraud: can external auditors spot

the difference? American Journal of Business and Management, 1 (1), 2012, 30-33

Kasznik, R., (1999). On the association between voluntary disclosure and earnings management.

Journal of Accounting Research, 37, 57-81.

Klein, A. (2002). Audit committee, board of directors‟ characteristics and earning management.

Journal of Accounting and Economics, 33, 375-400.

Klein, A. (2006). Audit committee, board of directors‟ characteristics and earning management.

Law & Economics Research Paper Series, Working Paper, 06-42.

Kothari, S. P., Leone, A. J., & Waslwy, C. E. (2005). Performance matched discretionary accrual

measures. Journal of Accounting and Economics, 39, 163-197.

Krishnan, G. V., & Parsons, L. M. (2005). Getting to the bottom line: an exploration of gender

and earnings quality. Journal of Business Ethics, 78, 65-76.

Lawler, E. E., & Conger, J. A. (2001). Strategy and business. 10(3), 431.457.

91
Lee, Y. (2013). Can independence directors improve the quality of earnings? Evidence from

Taiwan. Advance in Management and Applied Economics, 3(3), 45-66.

Lei, K. (2003). Earnings management and corporate governance in UK: the role of board of

directors and audit committee.

Lev, B. (1989). On the usefulness of earnings and earnings research: Lessons from two Decades

of empirical research. Journal of Accounting Research, 27, 153-201.

Lin, J. W. & Hwang, M. I. (2010). Audit quality, corporate governance, and earnings

management: a meta-analysis. International Journal of Auditing, 14 (1), 57-77.

Lipton, M., & Lorsch, J. W. (1992). Modest proposal for improved corporate governance. A

Business Law, 48(49).

Liu, S. (2011). Board governance and earnings management of Chinese listed companies. Master

thesis, universiteteti, Oslo.

Lorca, C., Sanchez-Ballesta, J., & Garcia-Meca, E. (2011). Board effectiveness and cost of debt.

Journal of Business Ethics, 100, 613–631.

Man, C. (2012). Corporate governance and earnings management: a survey of literature. Working

Paper.

92
Man, C., & Wong, B. (2013). Corporate governance and earnings management: a survey of

literature. Journal of Applied Business Research, 29(2).

Matsumoto, D. A. (2002). Management incentives to avoid negative earnings surprises. The

Accounting Review, 77, 483-514.

Metawee, A., (2013), “The relationship between characteristics of audit committee, board of

directors and level of earning management, Evidence from Egypt”, Journal of

International Business and Finance, Plymouth Business School, UK, January.

Modugu, K. P., & Dabor, E. L. (2013). Discourses on corporate governance and earnings

management: A theoretical approach. Journal of Accounting and Finance, 3(1), 31-

33

Moradi, M., Salehi, M., Bighi, S. J. & Najari, M. (2012). A study of relationship between board

characteristics and earnings management: Iranian scenario. Universal Journal of

Management and Social Sciences, 2(3).

Musa, I. F., Ibikunle, J., & Victor, C. O. (2013). Corporate governance mechanisms and reported

earnings quality of listed Nigerian insurance firms. International Journal of Finance

and Accounting, 2(5), 279-286.

Nahandi, Y. B., Baghbani, S. M., & Bolouri, A. (2011). Board combination and earnings

management: evidence from Iran. Journal of Basic and Applied Scientific Research,

1(12).
93
Nat OFO, (2013). Historical development of corporate governance in Nigeria: Corporate

governance, law and business. Available at http:// the corporateprof.com.

Niskanen, J., Karjalainen, J., Niskanen, M., & Karjalainen, J. (2009). Auditor gender and

corporate earnings management behaviour in private Finnish firms. Available at

ssrn.com/abstract.

Ntim, C. G., & Osei, K. A. (2011). The impact of board meetings on corporate performance in

South Africa. African Review of Economics and Finance, 2(2).

O‟Reilly, C. A., & Main, B. G. M. (2012). Women in the boardroom, symbol or substance?

Research Paper Series (2098), Standford graduate school of business.

Odia, J. O., & Ogiedu, K. O. (2013). Corporate governance, regulatory agency and creative

accounting practices in Nigeria. Mediterranean Journal of Social Sciences, Italy,

4(3).

OECD (1999). Principles of corporate governance. Http://www,encycogov.com.

Omoye, A. S., & Eriki, P. O. (2014). Corporate governance determinants of earnings

management: evidence from Nigerian quoted companies. Mediterranean Journal

of Social Sciences, Rome-Italy, 5 (3)

Onyeweaku, O. (2009). Cadbury pin revival on right issue-business hallmark, Proshare.

94
Oscar, N., & Daniel, V. (2013). Individual characteristics and earnings quality: executive

reputation and board director gender. Master’s Thesis in Accounting and

Financial Management, Department of Accounting, Stockholm School of

Economics

Oyejide, T. A., & Soyibo, A. (2001). Corporate governance in Nigeria. Paper presented at the

conference on corporate governance, Accra Ghana, 29-30 January, 2011.

Peasnell, K. V., Pope, P. F., & Young, S. (2000). Detecting earnings management using cross-

sectional abnormal accruals models. Lancaster University.

Pierce, J., & Zahra, S. (1991). Board composition from a strategic contingency perspective.

Journal of Management Studies, 29, 411-438.

Rafik, Z. A. (2002). Determinants of earnings management ethics among accountants. Journal of

Business Ethics, 40, 33-45.

Ramscar, M. (2002). Surveys, questions and other cheap ways of getting data.

Raut, S. (2010). Corporate governance-concepts and issues. India.

Riahi, Y., & Ben Arab, M. (2011). Disclosure frequency and earnings management: An analysis

in the Tunisian context. Journal of Accounting and Taxation, 3(3), 47-59.

95
Roodposhti, F. R., & Chasmi, S. A (2011). Impact of corporate governance mechanism on

earnings management. African Journal of Business Management, 5(11), 4143-4151.

Salah, A. (2010). Earnings management in the years following the integrated corporate income

tax within Dutch Housing Association. Erasmus University, Rotterdam.

Saleh, N. M., Iskandar, T. M., & Rahmat, N. M. (2005). Earnings management and board

characteristics: Evidence from Malaysia. Journal Pengurusan, 24, 77-103.

Schipper, K. (1989). Commentary on earnings management. Accounting Horizons, 91-102.

Scott, W. (2003). Financial accounting theory. Pearson Education, Toronto, Ontario

SEC (2003). The code of best practices on corporate governance in Nigeria.

SEC (2011). The code of best practice on corporate governance for public companies.

Shehu, U. H., & Abubakar, A. (2012). Ownership structure and opportunistic accounting: A case

of listed foods and beverage firms in Nigeria. International Journal of Physical and

Social Science, 2(7), 236-256.

Shehu, U. H. (2013). Financial reporting quality: Does monitoring characteristics matter? The

Business and Management Review, 3(2).

Short, Zhang, & Keasly (2002). The link between dividend policy and institutional ownership.

Journal of Corporate Finance, 8, 105-122.

96
Soliman, M. M., & Ragab, A. A. (2013). board of directors‟ attributes and earnings management.

Proceeding of 6th international business and social sciences research conference.

Dubai, UAE, 3-4 January, 2013.

Springer, K. (2008). Gender Diversity in Corporate Governance and Top Management, Journal

of Business Ethics 81, 83 – 95

Srindhi, B., Gul, F. A. & Tsui, J. (2011). Female directors and earnings quality. Contemporary

Accounting Research, 28(5), 1610-1644.

Sukeecheep, S., Yarram, S. R., & Al-farooque, O. (2013). Earnings management and board

characteristics in Thai listed companies. International Conference on Business,

Economics and Accounting, Bangkok, Thailand.

Sun, J., Liu, G., & Lan, G. (2011). Does female directorship on independent audit committees

constrain earnings management? Journal of Business Ethics, 99(3), 369-382.

Teoh, S. H., Welch, I., & Wong, J. J. (1998). Earnings management and the post issue

underperformance of seasoned equity offerings. Journal of Economic Economics,

50.

Trueman, B., & Titman, S. (1988). An explanation for accounting income smoothing. Journal of

Accounting Research, 26, 127-143.

97
Vafeas, N. (1999). Board meeting frequency and firm performance. Journal of Financial

Economics, 53.

Vincent, O. M., Olaegbe, B. A., & Sobona, A. L. (2008). Fundamental of business research. Olas

venture, Lagos, Nigeria.

Water, J. (2014). Correlational research design guidelines: Conducting correlational research.

Available at www.capilanou.ca.

Watts, R. L., & Zimmerman, J. L. (1986). Positive accounting theory. Englewood cliffs, N.J

Prentice- Hall.

Wei, Z., & Xie F. (2010). CFO gender and earnings management: Evidence from China.

Http://www.fma.org/Hamburg/Papers/wei_xie_efma2010.pdf.

Wolfensohn, J. (1999). Corporate governance is about promoting corporate fairness,

transparency and accountability. Financial Times, 21st June

Www.uky.edu/Classes/PSY/430-001/Correlational%20Research.ppt

Xie, B., Davidson, W., & Dadalt, P. (2003). Earnings management and corporate governance the

role of the board and audit committee. Journal of Corporate Finance, 9(3), 295-317.

98
Yousef, I. S. A., Nur Hidayah, B. L. & Kharil Faizal, B. K. (2014). Board of directors and

earnings management among Jordanian listed companies: proposing conceptual

framework. International Journal of Technical Research and Application, 2 (3), 01-07

Yusof, R. M. D. (2009). Malaysian code of corporate governance: the impact on the quality of

reported earnings of Kuala Lumpur composite index components.

Yusouff, W. F. W., & Armstrong, A. (2012). What competencies should directors possess?

Malaysian perspective.

Zhou, J., & Chen, K. (2004). Audit committee, board characteristics and earnings management

by commercial banks.

Appendix 1: Listed companies under foods, beverages, tobacco and breweries firms as at
2012

Beverages- Brewers/distillers:
1) Champion Breweries plc.
2) Golden Guinea Breweries plc.
3) Guiness Nigeria plc.
4) International Breweries plc.
5) Jos International Breweries plc.
6) Nigerian Breweries plc.
7) Premier Breweries plc.
Beverages- Non-alcoholic
8) 7-up Bottling Company plc.
Foods Products
9) Big Treat plc.
10) Dangote Flour Mills plc.
11) Dangote Sugar Refinary plc.
12) Flour mills of Nigeria plc.
13) Honey well Flour Mills plc.
14) Multi-Trex Integrated Foods plc.
15) Northen Nig. Flour Mills plc.
16) National Salt Company of Nig. Plc.
99
 17) P. S. Mandrides plc.
18) UTC Nigeria plc.
19) Union Dicon Salt
Foods Products Diversified
20) Cadbury Nigeria plc.
21) Nestle Nigeria plc.

Sources: NSE factbook, 2012

SAMPLED COMPANIES
1) Guiness Nigeria plc
2) Dangote Flour Mills plc
3) Flour mills of Nigeria plc
4) Nestle Nigeria plc
5) Nigerian Breweries plc.
6) Dangote Sugar Refinary plc.
7) Cadbury Nigeria plc.
8) National Salt Company of Nig. Plc.
9) Champion Breweries plc

APPENDIX II: Data for the study

Where DAC= Discretionary accruals extracted for each firm
year DAC BC FBM GM ROA FSIZE
2007 0.050742 0.333333 1.386294 0.066667 0.198869 18.08578
2008 0.008222 0.307692 1.386294 0.076923 0.204546 18.12556
2009 0.088757 0.357143 1.386294 0.071429 0.276385 18.08748
2010 0.113998 0.230769 1.386294 0.076923 0.271458 18.15375
2011 0.082507 0.333333 1.386294 0.25 0.287749 18.33977
2012 0.051097 0.333333 1.386294 0.25 0.222964 18.44571
2013 0.065843 0.333333 1.609438 0.25 0.171943 18.6118

2007 0.073881 0.444444 1.386294 0 0.018962 17.78073

2008 0.089306 0.4 1.386294 0 0.063871 17.87477
2009 0.048471 0.4 1.609438 0 0.134577 17.82201
2010 0.11526 0.4 1.386294 0 0.117406 17.90924
2011 0.081067 0.4 1.609438 0 0.06238 18.04651
2012 0.029329 0.4 1.609438 0 0.07332 17.98023
2013 0.030475 0.4 1.791759 0 0.067669 18.01392
0.001788 0.076923 1.386294 0 0.142064 17.90325

100
2007
2008 0.016105 0.083333 1.386294 0 0.103602 18.11194
2009 0.003348 0.076923 1.386294 0 0.125031 18.2501
2010 0.054184 0.142857 1.609438 0 0.165346 18.57538
2011 0.004177 0.142857 1.791759 0 0.109432 18.96596
2012 0.036485 0.142857 1.609438 0 0.117186 18.78962
2013 0.002554 0.214286 1.791759 0 0.088981 19.22666

2007 0.068287 0.111111 1.609438 0 0.39507 16.87198

2008 0.1097 0.25 1.386294 0 0.408224 17.18829
2009 0.0173 0.2 1.609438 0.1 0.355527 17.60537
2010 0.007066 0.222222 1.609438 0.111111 0.313742 17.91562
2011 0.041058 0.222222 1.386294 0.111111 0.279603 18.15861
2012 0.01876 0.222222 1.386294 0.111111 0.292138 18.30373
2013 0.039346 0.333333 1.609438 0.111111 0.260558 18.49956

2007 0.034592 0.214286 1.609438 0 0.302127 18.32139

2008 0.010128 0.214286 1.609438 0.071429 0.352236 18.46386
2009 0.021269 0.214286 1.609438 0.071429 0.389412 18.48823
2010 0.036345 0.230769 1.609438 0.076923 0.393101 18.55512
2011 0.012595 0.2 1.609438 0 0.289422 19.09839
2012 0.045098 0.230769 1.609438 0 0.252064 19.3514
2013 0.070588 0.230769 1.609438 0.076923 0.275846 19.34795
2007
0.216969 0.444444 1.609438 0.111111 0.611737 17.73001
2008 0.253732 0.444444 1.609438 0.111111 0.518637 17.87894
2009 0.077895 0.444444 1.609438 0.111111 0.250542 18.17453
2010 0.173617 0.444444 0.693147 0.222222 0.259237 17.94738
2011 0.140713 0.444444 1.609438 0.111111 0.163397 18.05117
2012 0.15383 0.444444 1.609438 0.111111 0.212475 18.23497
2013 0.006267 0.4 2.079442 0.2 0.230731 18.28271

2007 0.044366 0.3 1.386294 0.1 0.063894 16.20609

2008 0.053948 0.3 1.609438 0.1 0.058387 16.53959
2009 0.03329 0.5 1.386294 0.1 0.110575 16.5982
2010 0.115828 0.5 1.386294 0.1 0.151534 17.05527
2011 0.13614 0.5 1.386294 0 0.1213 17.31153
2012 0.005691 0.444444 1.609438 0 0.081759 17.70776
2013 0.068875 0.444444 1.94591 0 0.078339 17.83077

2007 0.002859 0.333333 1.386294 0.111111 0.288881 15.62188

2008 0.006514 0.333333 1.609438 0.111111 0.249104 15.84997
2009 0.048323 0.333333 1.386294 0.111111 0.333187 15.91414
2010 0.01003 0.333333 1.791759 0 0.263016 15.87829

101
 2011 0.130782 0.4 1.386294 0.1 0.312401 16.12276
2012 0.055706 0.4 1.791759 0.1 0.377597 16.18478
2013 0.040438 0.444444 1.791759 0.111111 0.354008 16.25185

2007 0.037828 0.285714 1.386294 0.142857 -0.03 15.66821

2008 0.016836 0.428571 1.386294 0.142857 -0.12243 15.26619
2009 0.054191 0.285714 1.098612 0.142857 -0.19609 15.13085
2010 0.015564 0.285714 1.609438 0.142857 -0.16588 14.84584
2011 0.059259 0.428571 1.386294 0 -0.27964 15.75546
2012 0.030959 0.333333 1.609438 0 -0.30757 15.73232
2013 0.044339 0.3 1.386294 0 -0.30516 16.02792

BM Board
BC(Number) (number) GM(number) size)
5 4 1 15
4 4 1 13
5 4 1 14
3 4 1 13
4 4 3 12
4 4 3 12
4 5 3 12
4 4 0 9
4 4 0 10
4 5 0 10
4 4 0 10
4 5 0 10
4 5 0 10
4 6 0 10
1 4 0 13
1 4 0 12
1 4 0 13
2 5 0 14
2 6 0 14
2 5 0 14
3 6 0 14
1 5 0 9
2 4 0 8
2 5 1 10
2 5 1 9
2 4 1 9
2 4 1 9
3 5 1 9
102
3 5 0 14
3 5 1 14
3 5 1 14
3 5 1 13
3 5 0 15
3 5 0 13
3 5 1 13
4 5 1 9
4 5 1 9
4 5 1 9
4 2 2 9
4 5 1 9
4 5 1 9
4 8 2 10
3 4 1 10
3 5 1 10
5 4 1 10
5 4 1 10
4 4 0 8
4 5 0 9
4 7 0 9
3 4 1 9
3 5 1 9
3 4 1 9
3 6 0 9
4 4 1 10
4 6 1 10
4 6 1 9
2 4 1 7
3 4 1 7
2 3 1 7
2 5 1 7
3 4 0 7
3 5 0 9
3 4 0 10

103
APPENDIX III: Descriptive, correlation and regression results and robustness test

104
- preserve
xtset id year, yearly
panel variable: id (strongly balanced)
time variable: year, 2007 to 2013
delta: 1 year

. su dac bc fbm gm roa fsize, detail

dac

Percentiles Smallest
1% .0017881 .0017881
5% .003348 .0025536
10% .0062671 .0028588 Obs 63
25% .0168361 .003348 Sum of Wgt. 63

50% .0443659 Mean .0573733

Largest Std. Dev. .0527176
75% .0778952 .1538298
90% .1307823 .1736171 Variance .0027791
95% .1538298 .2169691 Skewness 1.567622
99% .2537323 .2537323 Kurtosis 5.672357

bc

Percentiles Smallest
1% .0769231 .0769231
5% .1111111 .0769231
10% .1428571 .0833333 Obs 63
25% .2222222 .1111111 Sum of Wgt. 63

50% .3333333 Mean .3179468

Largest Std. Dev. .1130158
75% .4 .4444444
90% .4444444 .5 Variance .0127726
95% .4444444 .5 Skewness -.3739484
99% .5 .5 Kurtosis 2.245526
fbm

Percentiles Smallest
1% .6931472 .6931472
5% 1.386294 1.098612
10% 1.386294 1.386294 Obs 63
25% 1.386294 1.386294 Sum of Wgt. 63

50% 1.609438 Mean 1.524859

Largest Std. Dev. .1999211
75% 1.609438 1.791759
90% 1.791759 1.791759 Variance .0399684
95% 1.791759 1.94591 Skewness -.7046745
99% 2.079442 2.079442 Kurtosis 6.965723

gm

Percentiles Smallest
1% 0 0
5% 0 0
10% 0 0 Obs 63
25% 0 0 Sum of Wgt. 63

50% .0769231 Mean .0710594

Largest Std. Dev. .071201
75% .1111111 .2222222
90% .1428571 .25 Variance .0050696
95% .2222222 .25 Skewness .7181503
99% .25 .25 Kurtosis 2.985633

roa

Percentiles Smallest
1% -.3075662 -.3075662
5% -.1960856 -.3051631
10% -.029997 -.2796404 Obs 63
25% .0817587 -.1960856 Sum of Wgt. 63

50% .2045456 Mean .1788379

Largest Std. Dev. .182617
75% .2894221 .3950697
90% .3775969 .4082239 Variance .033349
95% .3950697 .518637 Skewness -.7059915
99% .6117372 .6117372 Kurtosis 3.914156

fsize

Percentiles Smallest
1% 14.84584 14.84584
5% 15.62188 15.13085
10% 15.75546 15.26619 Obs 63
25% 16.53959 15.62188 Sum of Wgt. 63

50% 17.91562 Mean 17.54078

Largest Std. Dev. 1.153102
75% 18.30373 19.09839
90% 18.6118 19.22666 Variance 1.329645
95% 19.09839 19.34795 Skewness -.6709565
99% 19.3514 19.3514 Kurtosis 2.36423

105
. pwcorr dac bc fbm gm roa fsize, star (0.05) sig

dac bc fbm gm roa fsize

dac 1.0000

bc 0.4898* 1.0000
0.0000

fbm -0.2620* -0.0278 1.0000

0.0381 0.8289

gm 0.1793 0.2463 -0.2202 1.0000

0.1597 0.0517 0.0828

roa 0.2937* -0.1001 0.1608 0.2184 1.0000

0.0195 0.4350 0.2079 0.0856

fsize 0.0616 -0.2907* 0.2037 -0.1334 0.4097* 1.0000

0.6315 0.0208 0.1092 0.2973 0.0009

. regress dac bc fbm gm roa fsize

Source SS df MS Number of obs = 63

F( 5, 57) = 10.50
Model .082612686 5 .016522537 Prob > F = 0.0000
Residual .08969404 57 .00157358 R-squared = 0.4795
Adj R-squared = 0.4338
Total .172306726 62 .002779141 Root MSE = .03967

dac Coef. Std. Err. t P>|t| [95% Conf. Interval]

bc .2721304 .0479924 5.67 0.000 .1760274 .3682335

fbm -.0943354 .0267012 -3.53 0.001 -.1478038 -.0408671
gm -.0797211 .0793498 -1.00 0.319 -.2386164 .0791741
roa .1090969 .0321817 3.39 0.001 .0446542 .1735396
fsize .0061677 .005104 1.21 0.232 -.0040529 .0163884
_cons -.007334 .0971094 -0.08 0.940 -.2017922 .1871242

. hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

Ho: Constant variance
Variables: fitted values of dac

chi2(1) = 11.48
Prob > chi2 = 0.0007

. vif

Variable VIF 1/VIF

fsize 1.36 0.732710

roa 1.36 0.734847
gm 1.26 0.795121
bc 1.16 0.862730
fbm 1.12 0.890670

Mean VIF 1.25

106
107
. xtset id year, yearly
panel variable: id (strongly balanced)
time variable: year, 2007 to 2013
delta: 1 year

. xtreg dac bc fbm gm roa fsize, fe

Fixed-effects (within) regression Number of obs = 63

Group variable: id Number of groups = 9

R-sq: within = 0.4590 Obs per group: min = 7

between = 0.0719 avg = 7.0
overall = 0.1416 max = 7

F(5,49) = 8.31
corr(u_i, Xb) = -0.7298 Prob > F = 0.0000

dac Coef. Std. Err. t P>|t| [95% Conf. Interval]

bc .0805772 .1024266 0.79 0.435 -.1252567 .2864111

fbm -.095275 .0236315 -4.03 0.000 -.1427643 -.0477856
gm -.2242518 .0829644 -2.70 0.009 -.3909751 -.0575285
roa .2844756 .0622397 4.57 0.000 .1594002 .409551
fsize .0212137 .0146083 1.45 0.153 -.0081429 .0505702
_cons -.2300092 .2338379 -0.98 0.330 -.6999241 .2399057

sigma_u .06089806
sigma_e .03220963
rho .78140481 (fraction of variance due to u_i)

F test that all u_i=0: F(8, 49) = 4.68 Prob > F = 0.0003

. est store fixed

. xtreg dac bc fbm gm roa fsize, re

Random-effects GLS regression Number of obs = 63

Group variable: id Number of groups = 9

R-sq: within = 0.3989 Obs per group: min = 7

between = 0.5030 avg = 7.0
overall = 0.4429 max = 7

Random effects u_i ~ Gaussian Wald chi2(5) = 40.56

corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000

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

bc .2403703 .0617739 3.89 0.000 .1192957 .3614449

fbm -.0893233 .0237676 -3.76 0.000 -.135907 -.0427396
gm -.1650623 .0806802 -2.05 0.041 -.3231926 -.0069319
roa .1616253 .0402104 4.02 0.000 .0828143 .2404362
fsize .0012809 .0067467 0.19 0.849 -.0119424 .0145042
_cons .0775102 .1186812 0.65 0.514 -.1551006 .310121

sigma_u .01997388
sigma_e .03220963
rho .27774386 (fraction of variance due to u_i)

. est store random

. hausman fixed random

Coefficients
(b) (B) (b-B) sqrt(diag(V_b-V_B))
fixed random Difference S.E.

bc .0805772 .2403703 -.1597931 .0817018

fbm -.095275 -.0893233 -.0059517 .
gm -.2242518 -.1650623 -.0591895 .0193339
roa .2844756 .1616253 .1228504 .0475069
fsize .0212137 .0012809 .0199328 .0129571

b = consistent under Ho and Ha; obtained from xtreg

B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic

chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 23.59
Prob>chi2 = 0.0003
(V_b-V_B is not positive definite)

108
Appendix IV: Fixed Effect Least Square Dummy Variables Regression
regress dac bc fbm gm roa fsize d1 d2 d3 d4 d5 d6 d7 d8

Source SS df MS Number of obs = 63

F( 13, 49) = 9.01
Model .121471173 13 .009343936 Prob > F = 0.0000
Residual .050835553 49 .00103746 R-squared = 0.7050
Adj R-squared = 0.6267
Total .172306726 62 .002779141 Root MSE = .03221

dac Coef. Std. Err. t P>|t| [95% Conf. Interval]

bc .0805772 .1024266 0.79 0.435 -.1252567 .2864111

fbm -.095275 .0236315 -4.03 0.000 -.1427643 -.0477856
gm -.2242518 .0829644 -2.70 0.009 -.3909751 -.0575285
roa .2844756 .0622397 4.57 0.000 .1594002 .409551
fsize .0212137 .0146083 1.45 0.153 -.0081429 .0505702
d1 -.136231 .0604331 -2.25 0.029 -.2576759 -.0147862
d2 -.1122855 .0455965 -2.46 0.017 -.2039152 -.0206559
d3 -.1631294 .0715182 -2.28 0.027 -.3068506 -.0194082
d4 -.1754228 .0639961 -2.74 0.009 -.3040278 -.0468179
d5 -.2034853 .0762298 -2.67 0.010 -.3566749 -.0502957
d6 -.0757234 .0561364 -1.35 0.184 -.1885337 .0370869
d7 -.0899141 .0351169 -2.56 0.014 -.1604842 -.0193441
d8 -.1337638 .0380193 -3.52 0.001 -.2101664 -.0573612
_cons -.1089031 .1951465 -0.56 0.579 -.5010646 .2832585

109