1

NDIC QUARTERLY SEPT & DEC,2018 VOL. 33 Nos 3&4

NDIC QUARTERLY

EDITORIAL BOARD

Editor-in-Chief S. A. Oluyemi, Ph.D.

Managing Editor K.S. Katata, Ph.D.

Members

Prof. Umar Sanda Prof. Bello Sabo

Prof. Ode Ojowu J. A. Afolabi, Ph.D. J. G. Donli, Ph.D. Tahir Sulaiman, Ph.D. K. O. Nwaigwe H. I. Ahmad

Associate Editors

T.W.O Alasia, Ph. D. H. Ochonogor, Ph.D. A. Abduraheem, Ph.D. F. Ogbeide, Ph.D. T. A. Aderemi, Ph.D. I. S. Alley, Ph.D. B. T. Mande, Ph.D. I. M. Idris (Secretary)

Editorial Secretariat

H. Sampson (Head) G. U. Muhammad

Y. Haliru, Ph.D. F. Imam

This Journal is a Quarterly Publication of the Nigeria Deposit Insurance Corporation (NDIC). Enquiries should be addressed to the Editor, NDIC Quarterly, Research,Policy and International Relations Department, Nigeria Deposit Insurance Corporation, P.M.B 284, Abuja, Nigeria. Electronic submissions are encouraged and should be sent to editor.quarterly@ndic.gov.ng

The views expressed in these articles are those of the authors and do not necessarily represent the official position of the NDIC. This Journal is solely financed by the NDIC towards the advancement of knowledge in Economics, Finance, Banking and Deposit Insurance practices. © Copyright and Published by the Nigeria Deposit Insurance Corporation

ISSN: 2636 7203 (Online)

2

Vision

To become one of the leading Deposit Insurers in the world.

Mission

To protect depositors and contribute to the stability of the financial system through effective supervision of insured institutions, provision of financial and technical assistance to eligible insured institutions, prompt payment of guaranteed sums and

orderly resolution of failed insured financial institutions.

Core Values

In its commitment to the public service, the NDIC employees have a tradition of

distinguished public service guided by the following six core values:

Honesty

Respect & Fairness

Discipline

Professionalism and Team work

Passion

3

NDIC QUARTERLY VOLUME 33 SEPT/DEC, 2018 NOS:3&4

TABLE OF CONTENTS

Content Page No 1. Review of Developments in Banking and Finance for Q3 and Q4 2018 By Research, Policy & International Relations Department

2. Financial Condition and Performance of DMBs for Q3 and Q4 2018 By Research, Policy & International Relations and Insurance & Surveillance Departments 3. MODERATING EFFECT OF AUDIT COMMITTEE ON BOARD DIVERSITY AND

EARNINGS MANAGEMENT IN NIGERIAN BANKS

By Musa Adeiza Farouk and Muhammad Aminu Isa Abstract

This study examines the impact of board diversity and audit committee on earnings management of listed deposit money banks and the effects of moderating factors that overcome the tendency of managers to engage in earnings manipulation. Using a population of fifteen listed deposit money banks in Nigeria, secondary data were obtained from the annual reports and accounts of the banks for the period 2008-2015. Multiple regression technique was adopted and Stata 13 used as the tool of data analysis. The findings revealed that before moderation all the variables, except board size, have significant effect on earnings management of banks. After moderation, the findings revealed that explanatory variables explained the extent of earnings management better than before moderation. The study recommends that the percentage of women director, shares held by directors and number of foreign directors should be increased, while the number of non-executive directors on audit committee should also be improved upon in order to mitigate the tendencies for manipulation of earnings management in banks. This study claims originality in the use of audit committee as moderating variable and the test of applicability and the usefulness of Chang, Shen and Fang (2008) model in Nigerian Banking Sector. Keyword: Earnings management, Audit Committee, Board Diversity, Echelon theory

4

4. Public Spending, Fiscal Sustainability and Macroeconomic Performance in

Nigeria

By Olufemi Muibi Sabiu

Abstract

This paper examines fiscal sustainability in Nigeria, vis-à-vis the economic performance in the country. Using the framework of an intertemporal budget constrain for the government, a fiscal sustainability equation is derived and the conditionality for establishing sustainability is ascertained. The empirical strategy applies the unit root test, cointegration test and dynamic OLS (DOLS) regression approach for testing the sustainability of the fiscal stance from 1961 to 2016. The empirical evidence shows evidence of weak sustainability especially as reported in the DOLS regression result. Similarly, the result for the effect of fiscal sustainability and economic performance also reports weak response of economic performance to fiscal sustainability. On the overall, the evidence from this study does not significantly deviate from extant studies in this strand of the literature. The main policy implication of this research is that the Nigerian government should ensure a more robust and systemic link between tax and expenditures policies and the evolution of public debt. In passing, a focus on determining a short-term government constrain framework and fiscal sustainability indicators for signaling short and medium term fiscal imbalances and to correct them will be a worthwhile direction for future research.

Keywords: Public Debt, Fiscal deficit, Fiscal Sustainability, Economic Performance,

intertemporal budget constraint

5. Capital Structure and Performance of Deposit Money Banks in Nigeria By Hafiz Usman Ahmed, Dr. Samaila I. Ningi & Dr. B.S. Dalhat Abstract Capital structure is one of the core decision areas in the field of finance, as it determines the existing amount of debt and equity of a bank. It is an important decision that has a close relationship with the value of bank hence its performance. Accordingly, the general objective of this study is to assess the impact of capital structure on the financial performance of Bank in Nigeria with specific reference to how debt ratio and equity ratio affect return on equity and net interest margin of banks in Nigeria. The population of the study is the entire 21 licensed DMBs in Nigeria (CBN, 2017). The sample size of 12 banks was determined using convenience sampling technique for the period 2007- 2016. The study utilizes panel design to analyse the data based on random effect estimation. The study found a positive relationship with financial performance measured by Net Interest Margin (NIM). The study recommends that more incentives need to be given to STD suppliers to effectively adjust the maturity structure of STDs. Similarly, debt should be used with caution in order to explore its tax shield and managerial efficiency benefits. Keywords: Capital Structure, Financial Performance, Deposit Money Banks

5

6. Effects of Switching Costs on Consumer Behavioural Intention in the Nigerian Deposit Money Bank

By: Ahmed Audu Maiyaki, Ph.D.

Abstract

The study investigates the moderating effect of switching costs on the relationship

between consumer behavioural intention and its antecedents such as functional

quality and technical quality with regards to retail banks in Nigeria. A sample of 800

bank customers was drawn from the selected banks operating in Kano metropolitan

city of Nigeria. A multigroup analysis was employed using Analysis of Moment

Structure (AMOS) software. The results revealed that switching costs have a

significant moderating effect on the paths between technical quality, and behavioural

intention. In contrast, switching costs do not have any significant moderating

influence on the paths between functional quality and behavioural intention. As

managerial and policy recommendations, managers should emphasise more on “what”

is actually delivered to customers rather than the “how” or the process of delivery.

Similarly, bank policy makers should deliver their services appropriately in such a way

that customers perceive greater benefits than the sacrifices they made.

Keywords: Behavioural intention, Swicthing cost, Functional quality, Technical

quality.

6

REVIEW OF DEVELOPMENTS IN THE THIRD AND FOURTH QUARTERS OF 2018

BY

RESEARCH, POLICY & INTERNATIONAL RELATIONS DEPARTMENT

1.0 INTRODUCTION

The Nigerian economy has experienced significant developments in Q3 and Q4 2018.

Some of these developments were outcomes of the contractionary monetary policy stance

maintained by the Central bank of Nigeria (CBN) Monetary Policy Committee (MPC)

meetings, since the July 2016 meeting. The highlights of global economic events are

discussed in sections 2 while sections 3-8 present significant developments in the Nigerian

economy.

2.0 THE GLOBAL ECONOMY

The global economy was estimated by the International Monetary Fund (IMF) to have

grown by 3.1% in Q4 2018, the same as Q3 2018 but lower than Q2 2018. In the G7

countries, the US experienced the highest growth due to increased consumer confidence,

healthy employment growth and strong wage growth. Japanese growth, in Q3 2018, was

hampered by natural disasters, while Canada’s growth momentum was inspired by a

strong labor market in the presence of lower oil prices. The Eurozone and the UK grew

at a slow rate due to social unrest in France, fragile drive from Germany and Brexit in the

UK. The Asian region, excluding Japan, as emerging market, had a robust growth due to

strong wage gains and labor markets, though impacted by weakened Chinese growth.

The IMF estimated, in its October 2018 World Economic Outlook, annual global economic

growth for 2018 at 3.7% in spite of notable headwinds such as weaker performance in

some economies, notably Europe and Asia, continued trade negotiations between the US

and China, the partial US government shutdown from 22nd December in Q4 2018 and

uncertainty regarding the Federal Reserve’s monetary policy that led to a tumultuous year

end for US in 2018.

3.0 DOMESTIC ECONOMY

The general economic performance in Nigeria improved as the growth in Gross Domestic

Product (GDP) rose from 1.5% in Q2 to 1.81% and 2.38% in Q3 and Q4 2018,

respectively. This indicates a rise of 0.55% compared with the growth rate recorded in

Q3 2018. The growth in the Q4 2018 exceeded market expectations of 2.1% which is the

highest growth rate since Q3 of 2015 as the non-oil sector increased further and

contraction in the oil sector contracted less. The oil sector shrank 1.6% in Q4 of 2018,

contributing 7.06% of real GDP in Q4 2018 compared to 9.38% in Q3 2018 and 8.55%

in Q2 2018. On the other hand, the non-oil sector contributed 92.94% to aggregate GDP

7

in Q4 2018, compared to 90.62% in Q3 2018 and 91.45% in Q2 2018. This improvement

in the non-oil sector was attributed to the increased performance in transport system,

electricity, information and communication, arts and entertainment etc.

The contributions of Agriculture, Industries and Services to GDP in the final quarter of

2018 were 26.15%, 20.24% and 53.62% respectively. The contributions represent

sizable improvement in Q4, compared to Q3 when these sectors respectively contributed

29.15%, 21.97% and 48.97% of the GDP. It is noteworthy that the Services Sector

accounted for over half of the GDP, indicating that the Nigerian economy is showing

indicator of positive economic transformation evident in a well-diversified economy like

developed countries. Figure 1 shows the real GDP growth rate from Q1 2017 to Q4 2018.

Fig 1: Real GDP Growth-Rate

Source: NBS

4.0 INFLATION

The inflation rate is measured as a percentage change in the Consumer Price Index (CPI).

CPI tracks variations in prices of goods and services consumed by people over time for

daily living. Headline inflation slightly rose to 11.28% at the end of Q3 2018 from 11.23%

at the end of the Q2, and further rose to 11.44% at end of Q4. This was attributed the

rise in prices of selected food items, incidences of farmers/herders clash in some parts of

the country also affected agricultural activities adversely in the period, fuel and lubricants,

among others. Figure 2 below shows the composite inflationary trends from January 2017

to December 2018.

-0.91

0.72 1.17 2.11 1.95 1.5 1.81 2.38

-20-16-12

-8-4048

1216202428

Q1-17 Q2-17 Q3-17 Q4-17 Q1-18 Q2-18 Q3-18 Q4-18

NON-OIL GDP GROWTH OIL GDP GROWTH GDP GROWTH

8

Figure 2 INFLATION RATE (COMPOSITE)

Source: NBS

5.0 EXCHANGE RATE MOVEMENT

There was stability in the exchange rate of the Naira to the US dollar in the year 2018,

both at the inter-bank and BDC segments of the foreign exchange market. That sustained

stability was on account of continuous intervention by the CBN in the foreign exchange

market and activities at the Investors’ and Exporters’ (I&E) window, which boosted

liquidity, dampened demand pressure and enhanced confidence in the market. For

instance, the CBN signed a 3-year bilateral currency swap agreement with the People’s

Bank of China, worth 15 billion Yuan (N 720 billion), equivalent to US$ 2.5 billion to boost

liquidity and facilitate trade and investment.

The exchange rate of the Naira to the US dollar at the inter-bank segment was

N306.70/US$ in Q4 2018, a slight rise from Q3 and Q2 figures of N306.03/US$ and

N305.77/US$ respectively. Thus, the premium between the inter-bank and BDC rates was

18.20% in Q4 2018, a slight increase from the Q3 figure of 17.40% but lower than Q2

figure of 18.34%. Figure 3 presents the trend of Naira/US$ rate in both the inter-bank

and the Bureau De Change (BDC) segments of the foreign exchange market.

18

.72

17

.78

17

.26

17

.24

16

.25

16

.10

16

.05

16

.01

15

.98

15

.91

15

.90

15

.37

15

.13

14

.33

13

.34

12

.48

11

.61

11

.23

11

.14

11

.23

11

.28

11

.26

11

.28

11

.44

9

Fig 3: Exchange Rate Movement

Source: CBN

6.0 EXTERNAL RESERVE MOVEMENT

The total external reserves in the fourth quarter of 2018 was US$42.54 billion,

compared with US$42.61 billion in Q3 2018 and US$47.16 billion in Q2 2018. The

growth was attributed to relative peace in the Niger-Delta that led to improved

crude oil sales, sustained high crude oil prices and Eurobond proceeds, amongst

others. Figure 4 presents the trend of Nigeria’s Gross Official External Reserve

from end-January 2017 to end-December 2018.

Fig 4. GROSS EXTERNAL RESERVES

Source: CBN

0.00

100.00

200.00

300.00

400.00

500.00

600.00

Jan

-17

Feb

-17

Mar

-17

Ap

r-1

7

May

-17

Jun

-17

Jul-

17

Au

g-1

7

Sep

-17

Oct

-17

No

v-1

7

Dec

-17

Jan

-18

Feb

-18

Mar

-18

Ap

r-1

8

May

-18

Jun

-18

Jul-

18

Au

g-1

8

Sep

-18

Oct

-18

No

v-1

8

Dec

-18

N/U

S$

Inter-Bank BDC

28

,59

3

29

,99

0

29

,99

6

30

,74

9

29

,81

2

30

,34

1

30

,89

9

31

,27

9

33

,16

0

34

,33

3

38

,20

8

39

,35

3

41

,15

0

45

,27

7

46

,73

1

47

,43

8

46

,92

3

47

,15

8

45

,81

4

44

,60

7

42

,60

9

40

,65

1

43

,34

8

42

,59

5

US$

' MIL

LIO

N

10

7.0 CAPITAL MARKET DEVELOPMENT

The Nigeria Stock Exchange (NSE) All-Share Index (ASI) was 38,278.55 in Q2 2018 which

waned to 31, 430.50 in Q4 2018, lower than its Q3 2018 value of 32,766.37, representing

a loss of 4.08%. The Equities and Debts (aggregate) market capitalization was N21.9

trillion in Q4 2018, 1.8% below the Q3 2018 value of N22.3 trillion. Similarly, equity

market capitalization as at the end of Q4 2018 stood at N11.72 trillion, showing a decline

of 2.13% from N 11.97 trillion in Q3 2018. The gradual decline in both the performance

indicators of the Exchange could be attributed to the impact of the progressive monetary

policy normalization in some advanced economies and the sustained profit taking

activities of foreign investors arising from perceived political risk in the build-up to the

2019 General Elections. However stock market regulators remained optimistic that the

trend will reverse in the medium term, albeit gradually, given the current stability in the

foreign exchange market and the external reserves position, as well as continued

improvements in key macroeconomic indicators. Figure 5 below highlights trends of the

two performance indicators in recent times.

Source: CBN Q4 2018, Economic Report

8.0 MONEY MARKET DEVELOPMENT

The money market was largely stable in the fourth quarter of 2018. Liquidity was

sustained by inflow from fiscal injections, Federal Government (FGN) Bonds, Nigerian

Treasury Bills (NTBs) and maturing Central Bank of Nigeria (CBN) bills. Outflow, such as

sale of CBN bills, FGN securities and provisioning and settlement for foreign exchange

purchases, impacted on market liquidity. Overall, banks continued to access the intraday

and standing facilities window to meet their short-term liquidity needs during the review

quarters.

26,874.62 25,516.34

33,117.48 35,439.98

38,243.19 41,504.51

38,278.55

32,766.37 31,430.50

-

5,000.00

10,000.00

15,000.00

20,000.00

25,000.00

30,000.00

35,000.00

40,000.00

45,000.00

0

5

10

15

20

25

30

Q4-16 Q1-17 Q2-17 Q3-17 Q4-17 Q1-18 Q2-18 Q3-18 Q4-18

IND

EX

N b

illio

n

Figure 5 : Market Capitalization and All-Share Index of NSE

Market Capitalization (N trillion) All-Share Index (equities)

11

Interest rates reflected the liquidity condition in the money market during the review

period. Specifically, developments in banks’ deposit rates were mixed in the Q3 and Q4

of 2018, averaging 8.52% and 8.63% respectively. That represent an increase of about

1.29 %in Q4, compared of decline of 2.29% in Q3. The average prime lending rates

trended downwards between Q2 and Q3 as well as Q3 and Q4 by 2.74% and 0.54%,

respectively. In the interbank funds segment, the weighted average interbank rates at

Q4 of same year stood at 9.93%; an increase of about 67.74% when compared with

value in Q3 2018 of 5.92%. That was in sharp contrast to decline of about 54.25% from

Q2 to Q3. The maximum lending rates have continued to show a falling trend since Q2,

Q3 and Q4 of 2018, registering a decline of 15.91% and 0.74% in Q3 and Q4 respectively.

Some of these trends in Q4 2016 - Q4 2018 period are depicted in figure 6 below.

Source: CBN

9.0 CBN MONETARY POLICY

The Monetary Policy Committee (MPC) held six meetings during the period under review,

on 23rd and 24th of July, 24th and 25th September, and 21st and 22nd November, 2018.

The decisions of the Committee after reviewing developments in both domestic and

international environments from the three meetings are as follows:

i. Retention of the MPR at 14 per cent;

ii. Retention of the CRR at 22.5 per cent;

iii. Retention of the Liquidity Ratio at 30.00 per cent; and

iv. Retention of the Asymmetric corridor at +200 and -500 basis points around the

MPR.

The monetary policy stance of the CBN since July 2016 has been contractionary and

aimed at containing inflation on the back of substantial expansionary fiscal policy, high

cost of energy, anticipated spending in the run-up to Edl Kabir/Christmas festivities,

campaign-related spending towards the upcoming 2019 general elections, increased

harvests, and stable exchange rate. The MPC believed that that policy stance would boost

0

10

20

30

40

Q4-16 Q1-17 Q2-17 Q3-17 Q4-17 Q1-18 Q2-18 Q3-18 Q4-18

per

cen

t p

er a

nn

un

Figure 6: Selected DMBs Interest Rates (Average)

average term deposit Prime Lending Interbank Maximum Lending

12

investor confidence, promote foreign capital flows and further enhance exchange rate

stability.

10.0 CBN Circulars

During the period under review, the CBN issued a number of circulars and guidelines to

guide the operations of insured deposit-taking financial institutions. The highlights of

some of these circulars are presented below:

10.1 BPS/DIR/GEN/CIR/05//007. Amendment to the Regulatory Framework for

BVN Operation and Watch-list for the Nigerian Banking Industry.

Issued on 4th July, 2018, the circular informs all stakeholders that the provision of valid

court order is now one of the prerequisites for the release of BVN information to

applicants.

10.2 BPS/DIR/GEN/CIR/05/009. Shared Agency Network Expansion Fund

Initiative: Regulatory Data Rendition Requirements.

The Circular issued on July 5,2018 instructed all banks including Microfinance and Primary

Mortgage banks that appoint agents, Mobile Money Operators and Licensed Super Agents

to render daily returns though the Nigeria Interbank Settlement System (NIBSS) to the

CBN. This is to plot the growth and type of services being offered across the country

10.3 Guidelines for Assessing Real Sector Support Facility (RSSF) Through Cash

Reserve Ratio (CRR) and Corporate Bonds.

The CBN announced the Guidelines after the Monetary Policy Committee’s (MPC) meeting

held from 23rd to 24th July 2018. It introduced Differentiated Cash Reserve Requirement

(DCRR) Regime where DMBs interested in providing credit financing to greenfield and

expansion projects in the real sector may request for the release of funds from their CRR

to finance these projects. This is to incentivize DMBs to direct affordable long term bank

credit to the manufacturing, agriculture and such other sectors considered by CBN as

employment and growth stimulating.

10.4 TED/FEM/FRC/GEN/01/004. Introduction of the Revised Foreign Exchange

Manual.

The circular issued on 26th of July 2018 introduce the revised the Revised Foreign

Exchange Manual intended to streamline documentation requirement, enhance

transparency of transactions and engender compliance by stakeholders.

10.5 BPS/DIR/GEN/CIR/05/010. Circular on Nigeria Bankers’ Clearing

System Rule, 2018 (Revised).

13

On July 31st the CBN issued the Revised Nigeria Bankers’ Clearing System (NBCS) Rule to

provide for efficient operation of automated clearing system. Speedy and efficient

collection of cheques. It also prescribed appropriate standards for the use of the NBCS.

The effective date of the new rules was rescheduled for 1st September, 2018.

10.6 Guidelines on the Management of Investment Account Holders (IAH) for Non-

Interest Financial Institutions (NIFIs) in Nigeria.

In August, 2018, the CBN issued exposure draft of the Guidelines on the Management of

Investment Account Holders (IAH) for Non-Interest Financial Institutions (NIFIs)

operating in Nigeria. It provided the minimum standard to be met by NIFIs before they

can recognize Profit Sharing Investment Accounts deposits as risk absorbent and thus

deduct same from the computation of Risk Weighted Assets towards the calculation of

Capital Adequacy Ratio.

10.7 Guideline on the Practice of Smoothing the Profit Payout to IAH for NIFIs in

Nigeria.

The CBN, in August 2018, issued exposure draft of Guideline on the Practice of Smoothing

the Profit Payout to IAH for NIFIs operating in Nigeria in order to regulate and standardize

income smoothing practice as well as specify and explain the techniques permitted in

Nigeria.

10.8 Exposure Draft Guidance Note on Supervisory Review Process for NIFIs in

Nigeria.

The Draft Guidance Note on Supervisory Review Process for NIFIs in Nigeria was issued

in August, 2018. It specifies the role of NIFIs in the computation of internal capital

requirement and capital management as well as the role of CBN in ensuring the

maintenance of adequate capital for the level of risks the NIFIs are exposed to.

10.9 Guidance Notes on Disclosure Requirement to Promote Transparency and

Market Discipline for NIFIs in Nigeria.

The exposure draft issued by the CBN in August, 2018 provided guidelines aimed at

achieving transparency and promote market discipline by allowing market participants

particularly the Investment Account Holders (IAH) to access relevant, reliable and timely

information on risk exposures and risk management policies and procedures of a NIFI.

10.10 Guidance Notes on the Calculation of Capital Requirement for Operational Risk

of NIFIs in Nigeria.

Issued by the CBN in August 2018, the Guidelines outline two methods of calculating

operational risk capital charge: the Basic Indicator Approach (BIA) and The Standardized

14

Approach (TSA) and required the NIFIs to assess the correlations among the various

types of risks and identify their possible impact in terms of operational risk.

1011 PMS/DIR/GEN/CIR/01/003. Exposure Draft of Regulatory Requirements for

Non-bank Merchant Acquiring in Nigeria.

The circular dated 7th September 2018 was addressed to all DMBs, Mobile Money

Operators, Payment Solution Service Providers, Payment Terminal Services Providers and

other Service Providers.

It announces the release of the Exposure Draft of Regulatory Requirements for Non-Bank

Merchant Acquiring in Nigeria for review and comments before the deadline of 21st

September, 2018.

10.12 FRP/DIR/GEN/07/007. Circular to all Commercial, Merchant, and Non-

interest Banks on Redesigned Credit Risk Management System (CRMS):

Additional Guidelines for the Operation of the Redesigned CRMS.

The CBN issued the additional guidelines on 10th September, 2018, to address some grey

areas in the rendition of requirements to ensure full compliance with the operations of

the CRMS.

10.13 BPS/DIR/GEN/CIR/05/011. Circular to all DMBs, MFBs, Other Financial

Institutions, Mobile Money Operators, Development Finance Institutions,

Payment Service Providers and Other Stakeholders: The Regulation of Instant

(Interbank) Electronic Funds Transfer Service (EFTS) in Nigeria.

The circular issued on 13th September, 2018, sets out rules for the operation of EFTS

which stated the rights and obligations of the parties involved. The rules prescribe N10,

000 penalty for failed NIP transaction not reversed in to customers account within twenty

four hours. The rules become effective on 2nd October, 2018.

10.14 Circular to all DMBs, Accredited Cheque Printers, and NIBSS: Revised Nigeria

Cheque Standard (NCS) and Nigerian Cheque Printers Accreditation Scheme

(NICPAS).

The circular issued on 18th of September, 2018 introduced the Revised Nigeria Cheque

Standard (NCS) and Nigerian Cheque Printers Accreditation Scheme (NICPAS) which is

going to be implemented from 1st February, 2019.However the new and old standards

shall run concurrently for one and a half years and by 1st August, 2020 only cheques that

conform to the new standard shall be allowed in the Automated Clearing System.

10.15 INCLUSION OF FERTILIZER ON THE LIST OF ‘’NOT VALID FOR FOREIGN

EXCHANGE’’ LIST.

15

On 12th October, 2018 the CBN announced the inclusion of fertilizer on the restricted

items not valid for foreign exchange list. This was announced in a circular with reference

number TED/FEM/FPC/GEN/01/006 to all authorized dealers and general public.

10.16 CIRCULAR FPR/DIR/GEN/CIR/08/064 DATED 17TH OCTOBER, 2018. THE

REDESIGNED CRMS: RELEASE OF MONTH END UPDATE TEMPLATE (CRMS-

400D).

The CBN introduced additional feature/ template to specifically enable participating

institutions perform ‘’Month End Update’’ of outstanding balances on their CRM records –

CRM-400D. Banks were advised to download the uploaded off line Validator for

subsequent updates.

10.17 CIRCULAR NO FPR/DIR/GEN/CIR/07/016 TO ALL MICROFINANCE BANKS ON

THE REVIEW OF MINIMUM CAPITAL REQUIREMENT.

The CBN, in exercise of powers conferred to it by the Banks and Other Financial

Institutions Act, and in furtherance of its mandate to promote a sound financial system

in Nigeria, has On 22ND October, 2018 increased the minimum capital requirement of

Microfinance banks (MFBs). The minimum capital requirement of Unit MFB is now N200

Million; that of State MFB is N1Billion while that of National MFB is N5Billion.

10.18 NOTIFICATION FOR CHANGE OF NAME FROM BANKING AND PAYMENT

SYSTEM DEPARTMENT (BPSD) TO BANKING SERVICES DEPARTMENT

(BKSD).CICULAR NO.MPS/DIR/GEN/CIR/05/014 ON 23RD OCTOBER, 2018.

The CBN informed all Banks and Other Financial Institutions, and Federal and State MDAs

on the restructuring of the erstwhile Banking and Payment System Department (BPSD)

in to two departments: Banking Services Department (BKSD) and Payment Services

Management Department. The BKSD shall handle matters relating to Banking Services

such as Treasury Single Accounts, Real Time Gross Settlement System, account opening,

Local and Foreign Payments transactions, banks settlement/ reconciliation issues,

Government Securities, Federation Account, Letter of Credit and related services.

10.19 CIRCULAR ON CODE OF CORPORATE GOVERNANCE FOR OTHER FINANCIAL

INSTITUTIONS. REFERENCE NO. FPR/DIR/CIR/GEN/01/017 DATES 26TH

October, 2018.

The CBN issued Code of Corporate Governance in respect of six Other Financial

Institutions which include Microfinance banks, Primary Mortgage banks and Development

Finance Banks. Others are Mortgage Refinancing Companies, Finance Companies and

Bureau de Change. The Codes become effective from April, 2019.

10.20 CIRCULAR ON THE REVISED NIGERIA CHEQUE STANDARD (NCS) AND

NIGERIAN CHEQUE PRINTERS ACREDITATION SCHEME (NICPAS): SORT CODE

AND MICR REJECT FLAG.

16

Vide circular with reference number BKS/DSO/GEN/MPF/01/003 dated 7th

November,2018, the CBN notified all DMBs and NIBSS on the implementation of the NCS

and NICPAS aka Green Book Version 2.0 using the unique state codes stipulated as

components of the sort code on Cheque instruments. Each state has a unique code that

must be used by all branches operating within the state. The circular also required all

DMBs to activate the MICR reject/repair flag on their clearing system

10.21 EXPOSURE DRAFT FOR THE REVISED STANDARDS ON NIGERIA UNIVERSAL

BANK ACOUNT NUMBER (NUBAN) SCHEME FOR BANKS AND OTHER

FINANCIAL INSTITUTIONS. PMS/DIR/GEN/EXP/01/002 DATED 6TH

DECEMBER, 2018.

The CBN issued an exposure draft for the Revised Standards on Nigeria Universal Bank

Account Number (NUBAN) Scheme for Banks and Other Financial Institutions for review

and comments. The draft introduced new Cheque Digital Algorithm intended to include

Other Financial Institutions (OFI) who were expected to comment before 26th November,

2018.

10.22 FOREIGN EXCHANGE RESTRICTION ON THE IMPORTATION OF 42 ITEMS.

In a letter referenced FRR/DIR/GEN/CIR/07/019 dated 10th December, 2018, the

CBN observed the circumvention of the Foreign Exchange Restriction Policy by some

banks. Therefore, the CBN informed all DMBs that its Economic Intelligence Unit and the

Economic and Financial Crimes Commission (EFCC) would commence immediate

investigation of the accounts of corporates and entities engaged in this unwholesome act

with a view to visiting severe sanctions on all the culprits.

10.23 DEPLOYMENT OF CONSUMER COMPLAINTS MANAGEMENT SYSTEM (CCMS).

In a circular with reference number CPD/DIR/GEN/CIR/01/001 addressed to all

Banks and Other Financial Institutions on 21st December, 2018, the CBN announced the

deployment of the CCMS effective 2nd January, 2019. The system requires the institutions

to assign tracking number of every complaint received from their customer, issue

acknowledgement and commence upload of complaints to the CCMS on a daily basis.

17

FINANCIAL CONDITION AND PERFORMANCE OF DEPOSIT MONEY BANKS

IN THE THIRD AND FOURTH QUARTER OF 2018 By

Research, Policy & International Relations AND Insurance and Surveillance

Departments

1.0 INTRODUCTION The Nigerian economy continued to witness marginal recovery during the third and fourth quarters of 2018. On the other hand, the Consumer Price Index (year-on-year) which measures Inflation, rose from 11.28% in September 2018 to 11.44% in December 2018. Also, there was relative stability in both the Investors’ and Exporters’ (I&E) window of the foreign exchange market due to autonomous inflows and sustained economic intervention efforts of the CBN. However, the intervention resulted in the decline of Nigeria’s External Reserves which stood at US$47.80 billion in June 2018, but dropped to

US$43.11 billion in December 2018.

The CBN has continued to maintain MPR rate at 14% to contain inflation and achieve price stability. Unfortunately, these rates have remained high with the attendant adverse effect on liquidity positions and funds accessibility by the private sector. Also, the maximum lending rate has been hovering between 30% and 31% since December 2017, thus making access to finance really difficult for businesses especially the Micro, Small and Medium Enterprises (MSMEs). Nonetheless, with the recent special MSME intervention funds by the CBN, small businesses could have access to affordable financing which could boost the economy. These economic developments influenced financial sector performance in the third and fourth quarters of 2018, as reflected in key financial indicators which were mostly on an increasing trend. Total Industry Assets increased from ₦33.22 trillion in Q2 2018 to ₦34.84 trillion in Q3 2018 and marginally increased again to ₦35.10 trillion in Q4 2018. Total Deposits from Customers increased from ₦20.37 trillion in Q2 2018 to ₦20.92 trillion in Q3 2018 and further grew to ₦21.73 trillion in Q4 of 2018. The Capital to Risk-Weighted Assets Ratio (CAR) also improved from 12.08% in Q2 2018 to 14.14% in Q3 2018 and additionally increased to 15.27% in Q4 2018. The rest of this chapter is organized as follows. Section 2 presents the structure of Assets and Liabilities. Section 3 assesses the financial condition of insured banks, while Section 4 concludes. 2.0 STRUCTURE OF ASSETS AND LIABILITIES Total Industry Assets marginally increased from ₦33.22 trillion recorded in Q2 2018 to ₦34.84 trillion in Q3 2018. The trend continued in Q4 2018 as Total Industry Assets rose to ₦35.10 trillion. The increment was largely attributed to the increase in Amortized Cost by 35.95% from ₦1.89 trillion in Q3 2018 to ₦2.57 trillion in Q4 2018. Fair Value through Other Comprehensive Income (FVOCI), Assets Pledged as Collateral, and Financial Assets held for Trading also increased by 13.13%, 10.82% and 7.92%, respectively, during the

same period.

18

On the Liabilities side, Total Deposits from Customers which constituted 61.32% of the Industry Total Liabilities increased from ₦20.37 trillion in Q2 2018 to ₦20.92 trillion in Q3 2018 and grew further to ₦21.73 trillion in Q4 2018, which is an increment of 3.85%. Seven (7) banks had Total Deposits in excess of ₦1 trillion each. Cumulatively, these banks had ₦14.18 trillion or 65.25% of the Total Industry Deposits of ₦21.73 trillion in Q4 2018. The structure of the industry’s Total Assets and Liabilities for Q2 2018, Q3 2018 and Q4 2018 are presented in Table 1 and Charts 1.1 and 1.2. TABLE 1: Structure of DMBs’ Assets and Liabilities for Q3 and Q4 2018

ASSETS

Q2 2018

(% OF TOTAL

ASSETS)

Q3 2018

(% OF TOTAL

ASSETS)

Q4 2018

(% OF TOTAL

ASSETS)

Change

Cash Balances 2.19 2.15 2.25 5.46

Balances with Banks & Central Bank 23.61

23.36 23.19 0.02

Loans & Advances to

Banks 1.18 1.35 1.19 - 11.05

Loans & Advances to Customers 39.27

38.03 36.67 - 2.84

Financial Assets Held for Trading 4.92

4.57 4.89 7.92

Investment Securities:

Available for Sale 10.65 10.26 11.52 13.13

Investment Securities: Held to Maturity 5.28

5.42 7.32 35.95

Assets Pledged as Collateral 3.50

3.48 3.83 10.82

Investment in

Subsidiaries & Associates 1.22

1.16 1.24 7.83

Investment Properties 0.45

0.03 2.41 1,040.52

Property Plant and

Equipment 2.43 2.77 5.03 - 12.34

Other Assets 5.26

7.36 5.03 - 31.15

Asset Classifies as Held for Sale & Discontinued

Operations 0.05

0.05 0.07 38.39

TOTAL ASSETS 100 100.00 100.00 0.76

LIABILITIES % OF TOTAL LIABILITIES

% OF TOTAL LIABILITIES

% OF TOTAL LIABILITIES

Deposit from Banks 4.73 4.43 5.00 13.79

Deposit from Customers 61.32

60.06 61.90 3.85

Financial Liabilities Held for Trading 0.06

0.15 0.13 - 12.61

19

Borrowings 9.18 8.96 8.69 - 2.28

Debt Instrument 4.09 3.65 3.63 0.04

Other Liabilities 11.70 12.31 11.65 - 4.63

Shareholders' Fund 8.92 10.44 8.99 - 13.19

TOTAL LIABILITIES 100.00

100.00 100.00 0.76

Source: NDIC

CHART 1.1: Structure of DMBs Assets for Q3 and Q4 2018

CHART 1.2: Structure of DMBs Liabilities for Q3 and Q4 2018

Table 1 and chart 1 shows that Loans & Advances to Customers was the largest components of the Total Assets of DMB’s in both quarters accounting for 38.03% &

2.25

23.19

1.19

36.67

4.89

11.527.32

3.831.24 2.41

5.03 5.030.07

05

10152025303540

Q3 2018 (% OF TOTAL ASSETS) Q4 2018 (% OF TOTAL ASSETS)

5

61.9

0.13

8.693.63

11.65 8.99

0

10

20

30

40

50

60

70

Deposit fromBanks

Deposit fromCustomers

FinancialLiabilities Held

for Trading

Borrowings DebtInstrument

Other Liabilities Shareholders'Fund

% OF TOTAL LIABILITIES % OF TOTAL LIABILITIES

20

36.67% in Q3 and Q4 of 2018, respectively. Balances with Banks & Central Bank came

second with 23.36% and 23.19% in Q3 and Q4 2018, respectively.

Total Deposits from Customers was the largest component on the Liabilities side with 60.06% in Q3 2018 and 61.90% in Q4 2018. Other Liabilities and Shareholders’ Funds accounted for 11.65% and 8.99% in Q4 2018. 3.0 FINANCIAL CONDITION OF DMBs 3.1 Capital Adequacy The Capital to Risk-Weighted Assets Ratio (CAR) improved from 12.08% in Q2 2018 to 14.14% and 15.27% in Q3 and Q4 2018 respectively. The increase was attributable to the decline in Total Risk Weighted Assets from ₦21.49 trillion in Q3 2018 to ₦20.90 trillion in Q4 2018. Total qualifying Capital also increased from ₦3.04 trillion in Q3 2018 to ₦3.19 trillion in Q4 2018.

TABLE 2: DMBs Capital Adequacy Position for Q3 and Q4 2018

Capital Adequacy Ratio Q2 2018 (%) Q3 2018 (%) Q4 2018 (%)

Capital to Risk Weighted Assets 12.08% 14.14% 15.27%

Capital to Total Assets Ratio 7.78% 8.72% 9.09%

Adjusted Capital Ratio 15.44% 18.69% 20.10%

Source: NDIC

Despite the rise in the CAR, capital shortfalls of four (4) banks translated to a recapitalization requirement of ₦726.23 billion in Q3 2018 and, ₦704.88 billion in Q4 2018. This indicates a decreasing trend from Q2 2018 figure of ₦1.43 trillion. Table 2 and Chart 2 depict the CAR position of the industry for Q1 and Q2 2018.

CHART 2: DMBs Capital Adequacy for Q3 and Q4 2018

15.27%

9.09%

20.10%

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

Capital to Risk WeightedAssets

Capital to Total AssetsRatio

Adjusted Capital Ratio

Q3 2018 (%) Q4 2018 (%)

21

3.2 Asset Quality

The Banking Industry Total Earning Assets, which constituted 62.81% of the Industry Total Assets, increased by 4.43%, from ₦21.12 trillion Q3 2018 to ₦22.05 trillion in Q4 2018. Total Credit which represents the bulk of the Earning Assets, decreased slightly by 3.34% from ₦15.81 trillion in Q3 2018 to ₦15.29 trillion in Q4 2018.

Out of the Industry Total Credits (TCs) of ₦15.29 trillion, impaired Credits amounted to ₦1.79 trillion or 11.70%, indicating an improvement from Q3 2018 figure of ₦2.24 trillion, or 14.18% of Total Credits. The total credit was ₦15.52 trillion in Q2 2018, out of which ₦1.94 trillion or 12.47% was impaired. However, Impaired Credits continued to exceed the maximum threshold of 5% prescribed by the CBN. Credits to insiders amounted to ₦721.28 billion in Q3 2018, 50.79% of which or ₦366.31 billion was impaired. The figure rose to ₦756.40 billion in Q4 2018, 46.15% or ₦349.11 billion was impaired. Credit to the Oil and Gas sector remained dominant, standing at ₦4.87 trillion in Q3 2018 or 30.81% of Total Credits, out of which ₦966.86 billion or 43.50% was impaired. The figure declined slightly in Q4 2018, standing at ₦4.66 trillion or 30.46% of Total Credits. The impaired amount stood at ₦878.40 billion. Q3 2018 and Q4 2018 Asset Quality indicators are shown in Table 3 and Chart 3.

TABLE 3: DMBs Asset Quality Indicators in Q3 and Q4 2018

Details Q2 2018 (%)

Q3 2018 (%) Q4 2018 (%)

Impaired Credit to Total Credit 12.47 14.17 11.70

Total Earning Assets to Total Assets 59.74 60.61 62.81

Impaired Credit to Shareholders' Fund 86.66 77.27 57.5

Source: NDIC

CHART 3: DMBs’ Asset Quality for Q3 and Q4 2018

11.70

62.8157.5

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

Impaired Credit to Total Credit Total Earning Assets to Total Assets Impaired Credit to Shareholders'Fund

Q3 2018 (%) Q4 2018 (%)

22

3.3 Earnings and Profitability

The industry recorded a Profit Before Tax (PBT) of ₦306.48 billion in Q4 2018, which was much higher than the Q3 2018 figure of ₦122.57 billion and also greater than ₦159.51 billion of Q2 2018. That could be attributed to a 23.89%, 29.27% and 20.12% decrease

in Interest expense, Operating expenses and Trading Income, respectively.

Return on Assets (ROA) increased from 0.24% in Q3 2018 to 0.88% in Q4 2018. Also, Return on Equity (ROE) increased from 4.56% in Q3 2018 to 9.73% in Q4 2018. The Q3 2018 and Q4 2018 Earnings and Profitability indicators are shown in Table 4 and Charts

4.1 and 4.2.

TABLE 4: DMBs Earnings and Profitability Indicators in Q4 2017and Q1& Q2 2018

Indicators Q2 2018 (%) Q3 2018 (%) Q4 2018 (%)

Return on Assets 0.64 0.24 0.88

Return on Equity 6.65 4.56 9.73

Net Interest Margin 1.40 1.20 1.85

Yield on Earning Assets 2.54 2.46 3.23

Parameters Q2 2018 (N' billion)

Q3 2018 (N' billion)

Q4 2018 (N' billion)

Profit Before Tax 195.19 122.57 306.48

Interest Income 767.78 769.95 698.16

Operating Expenses 430.89 465.18 329.04

Interest Expense 343.58 392.85 298.98

Net-Interest Income 424.19 377.09 399.18 Source: NDIC

CHART 4.1: DMBs Earning and Profitability ratios for Q3 and Q4 2018

Source: NDIC

0.88

9.73

1.85

3.23

0

2

4

6

8

10

12

Return on Assets Return on Equity Net Interest Margin Yield on EarningAssets

Q3 2018 (%) Q4 2018 (%)

23

CHART 4.2: Earning and Profitability for Q3 and Q4 2018

Source: NDIC

3.4 Liquidity Profile

In Q2 2018, the Average Liquidity Ratio (ALR) of the industry was 61.75%. The ratio rose to 83.18% in Q3 2018 and then declined to 51.87% in the final quarter of 2018. The Net Credit to Deposits Ratio, measures banks’ level of lending activities, stood at 67.33% in Q3 2018 then declined slightly to 64.69% in Q4 2018, as shown in Table 5 and Chart 5. All except two (2) of the DMBs met the minimum liquidity ratios of 30% for Commercial Banks and 20% for Merchant Banks in Q4 2018.

TABLE 5: DMBs Liquidity Profile for Q4 2017 and Q1 2018

Indicators Q2 2018 (%) Q3 2018 (%) Q4 2018 (%)

Average Liquidity Ratio 61.75 83.18 51.87

Net Credit to Deposit Ratio 67.86 67.33 64.69

Inter-Bank Takings to Deposit Ratio 1.61 7.38 1.73

No of Banks with Liquidity Ratio below the prescribed minimum 2

2 2

Source: NDIC

CHART 5: DMBs Liquidity Profile for Q1 2018 and Q2 2018

306.48

698.16

329.04298.98

399.18

0

100

200

300

400

500

600

700

800

900

Profit Before Tax Interest Income OperatingExpenses

Interest Expense Net-InterestIncome

Q3 2018 (N' billion) Q4 2018 (N' billion)

24

4.0 CONCLUSION DMBs had an improved performance in the second half (Q3 and Q4) of 2018. During the quarters under review, the Banking Industry Average CAR increased from 14.14% in Q3 2018 to 15.27% in Q4 2018. Total Industry Assets increased by 0.76% from Q3 to Q4 2018. Profit Before Tax increased by 150.04%. Impaired Credits to Total Credits and Operating Expenses declined, while Recoveries also improved during the period

under review.

51.8764.69

1.730

20

40

60

80

100

Average Liquidity Ratio Net Credit to Deposit Ratio Inter-Bank Takings toDeposit Ratio

Q3 2018 (%) Q4 2018 (%)

25

MODERATING EFFECT OF AUDIT COMMITTEE ON BOARD DIVERSITY AND

EARNINGS MANAGEMENT IN NIGERIAN BANKS

By Musa Adeiza Farouk and Muhammad Aminu Isa

1 Introduction Corporate environment in Nigeria has experienced cases of earnings management, such as the reported manipulative accounting scandal in African Petroleum Plc, Cadbury Nigeria Plc in 2006 and also the case of Oceanic Bank Plc and Intercontinental Bank Plc. This has brought doubt on the credibility of financial reporting. Based on this series of reported accounting scandals, there arises the need to identify factors that could be used to mitigate the management tendencies to engage in manipulative accounting practices. Diversity of the board members is believed to be one of the factors that could be used to curtail the extent of manipulative accounting (Shehu & Farouk, 2014).

According to Marimuthu (2008), board diversity is the variation in age, race, ethnicity, gender, and social/cultural identities among employees within a specific corporation. These diversities have attracted researcher’s attention due to their expected effect on earnings management. Participation of women on boards of directors (gender diversity) is expected to influence the extent of earnings management because of their effectiveness in attendance of meeting and adherence to ethical codes and standards which they possess over their male counterparts on the board (Vafeas & Theodorou, 1998). Therefore, studies on ethics posit that women are less likely to engage in unethical

behavior at place of work in order to obtain financial rewards (Rose, 2007).

Also, the percentage of shares owned by board of directors is expected to significantly affect the extent of earnings management. It is either the percentage of shares enables them to align their interest with that of the owners thereby reducing agency conflict or that the level of shares held by them might make them become entrenched since certain level of shares may put the managers in a dominant position. This may permit exploitation of external minority shareholders. In addition, nationality of the board members and their stake in the organization could also serve as an incentive to monitor the management in

order to safeguard their investment (Hooghiemstra, Hermes, Oxelheim & Randoy, 2015).

Size and the composition of the board members was posited by Yu (2013) to be effective and efficient for board performance. Therefore, the size and the composition of the board are expected to serve as a driving force against manipulative activities of management in

the banks.

Consequently, audit committee has been identified as core to financial reporting quality. Meanwhile, the basic function of audit committees is to oversee the financial reporting process, monitor managers’ tendencies to manipulative earnings, increase the audit quality and to reduce the questioning of board of directors. Hence, there is an increasing advocacy for inclusion of women in audit committee because they are naturally less likely to engage or encourage manipulation. The assumption is that, if women serve on board and also in the audit committee, they may be able to mitigate the managers’ tendencies towards earnings management. Also, since the audit committee may be composed of

26

board members who have shares in the bank, it is expected that the multiple role of the board and the audit committee functions will affect the level of earnings management significantly. This justifies the use of audit committee to moderate the relationship

between women director, board ownership and earnings management.

Studies such as Gulzar and Wang (2011), Guo, Huang, Zhang and Zhou (2014), Arun, Almahrog and Aribi (2015), Hussaini and Gugong (2015), Ibrahim (2015) and Ishaq, (2017) explored the influence of board diversity on earnings management in both developed and developing countries, but none of these studies consider the use of moderating effect of one variable on the relationship between board diversity and earnings management. As such, this study introduces audit committee as a moderating variable to examine its multiplicative role on the relationship between board diversity and earnings management in listed Deposit Money Banks in Nigeria.

Most of the studies (such as Gulzar & Wang, 2011; Qi & Tian, 2012; Saleh & Haat, 2014; Baccouche, Hadriche & Omri, 2014) were conducted in developed countries, while few of them (such as Razek, 2012; Shehu & Ibrahim, 2014; Kantudu & Samaila, 2015; and Ishaq, 2017) were carried out in Nigeria. Therefore, there is the need to add to the existing literature in another dimension using moderating variable.

The study is motivated by the fact that, corporate governance culture ab-initio in Nigeria has consistently failed to be responsible and accountable to the stakeholders. It has no deep-rooted mechanism to maintain a balance among the major players which have resulted in increasing levels of manipulation of reported accounting numbers as posited by Bello (2005) cited in Shehu (2012). This study therefore seeks to establish the effect

of board diversity and audit committee on earnings management of banks in Nigeria.

The result would have an important policy implication for regulators who are striving to improve the transparency and quality of financial reporting in the financial sector especially in assessing the effectiveness and otherwise of the existing audit committee and the board in ensuring quality financial reports devoid of misstatement. Among the important policy implications is that, the study suggests the need by the management of the banks to encourage the full application of corporate governance codes in banks. However, this provides effective and efficient monitoring of financial reports most importantly the earnings, especially for those banks with high percentage of board

ownership and foreign directors.

The main objective of the study is to examine the impact of board diversity and audit committee on earnings management of listed Deposit Money Banks in Nigeria. The specific objective is: to determine the moderating effect of audit committee on the relationship between women directors, board ownership and earnings management of listed Deposit Money Banks in Nigeria.

In order to achieve the objective of the research, this paper is organized into five sections, with this section being the introduction. Section 2 is dedicated to review of relevant literature. Section 3 is devoted to the methodology used by the study. Section 4 is concerned with presentation, interpretation, analysis and discussion of results. Finally, section 5 offers a conclusion as well as a set of recommendations drawn from the findings

of the study.

27

2. Literature Review and Theoretical Framework 2.1 Review of Literature 2.1.1 Earnings Management: Earnings management is defined as the deliberate dampening of fluctuations about “some level of earnings considered being normal for the firm (Barnea, Ronen & Sandan, 1976). Also, Copeland (1968) defined earnings management to involve repetitive selection of accounting measurement or reporting rules in a particular pattern, which is aimed at reporting a stream of income with a smaller variation from a trend than would otherwise have appeared. For the purpose of this study, the definition given by Copeland (1968) is adopted for the study. 2.1.2 Board Diversity: Board diversity refers to the composition of the board by combining attributes, characteristics and expertise of individual board members that contribute to board processes and decision making in a positive way (Burton & Ryall, 1995). In addition, Mazur (2010) saw diversity as a subjective phenomenon, created by group themselves who on the basis of their different social identities categorize others as similar or dissimilar. 2.1.3 Audit Committee: Audit committee is a sub-committee which is created by the board of directors with the aim of making arrangements for the audit, carrying a regular check and overview of the financial statement. This committee enhances the ability of the board to fulfil its legal responsibilities and ensure that the credibility and objectivity

of the financial reports are attained (CBN code of corporate governance for banks, 2006).

2.2.1 Women Director and Earnings Management Applying panel least square regression to a set of data on a sample of listed manufacturing firms in Nigeria, Shehu and Ibrahim (2014) reported that women directors have significant positive impact on real activity manipulation. Similar results were reported by Oba (2014) based on multiple regression technique on Nigerian data. He found that gender diversity had significant negative impact on earnings management. This echoes the findings reported by Omoye and Eriki (2014) who applied probit, logit and extreme regression to a sample of 132 Nigerian firms and found a significant negative effect of board gender on earnings management. In France, Lakhal, Aguir, Lakhal and Malek, (2015) used multiple regression with 170 firms and found a negative relation between the proportion of women on board and earnings management of firms. Susanto (2016) using a sample of 69 firms from Indonesia showed that gender has influence on earnings management of their manufacturing firms. Also, Hussaini and Gugong (2015) in Nigeria used ordinary least square regression and found that female director had a significant positive influence on earnings management. Ioualalen, Khemakhem and Fontaine (2015) using multiple regression technique found that gender diversity have no significant influence on earnings management of firms for both Iranian and Canadian

firms.

2.2.2 Board Ownership and Earnings Management A study conducted by Kantudu and Samaila (2015), reported that managerial shareholding was significant in influencing financial reporting quality of quoted oil marketing firms in Nigeria. Ramadan (2015) revealed that management ownership was inversely associated with the practices of earnings management. Ogbonnaya, Ekwe and Ihendinihu (2016) reported that managerial ownership had positive significant effect on earning management of brewery industry in Nigeria. Fei (2015) used 323 sample firms and found that managerial ownership had positive and significant impact on earnings reliability implying that managerial ownership reduced the level of earnings management.

28

In Nigeria, Kurawa and Saheed (2014) used multiple regression technique and found that insider ownership and management equity holding had significant positive influence on

earnings management within the study period.

2.3.3 Foreign Director and Earnings Management In a study from Malaysia by Abdul Rauf, Johari, Buniamin and Abd Rahman (2012) using data from 214 companies revealed that board race did not influence the practice of earnings management. A study from Netherland by Hooghiemstra, Hermes, Oxelheim and Randoy (2015) found that foreign director participation tended to exhibit a significant positive relation with earnings management of firms. On the other hand, Zhang and Uchida (2011) and Lei (2003) found that foreign board membership and, independent

foreign institutional investors had significant impact on earnings management of firms.

2.3.4 Board Size and Earnings Management Board size does not influence the practice of earnings management according to the study of Abdul Rauf, Johari, Buniamin and Abd Rahman (2012). Using a sample of Turkish firms and multivariate analysis, Mustafa, Mehmet and Suleyman (2014) reported that board size had negative and significant effect on the earnings management. In a related study by Oba (2014), he documented negative but insignificant effect of board size on earnings management of firms in Nigeria. Fodio, Ibikunle and Oba (2013) used multiple regression technique and found that board size was negatively and significantly associated with earnings management. Holtz and Neto (2013) revealed that earnings information was negatively affected by board size (more than nine members) in listed Brazilian Firms. Salihi and Jibril (2015), studying a sample of 29 consumer good firms suggested that larger board was not efficient to minimize the tendency of managing earnings. Also, a study conducted by Patrick, Paulinus and Nympha (2015) using a simple regression showed that board size had significant influence on earnings management practices

among Nigerian quoted companies.

2.3.5 Board Composition and Earnings Management. A study by Zhang and Li (2007) in Chinese firms showed that the extent of earnings management was negatively related to the proportion of the independent directors. Similarly, Dimitropoulos (2011) adopted ordinary least squares regression and found that clubs with increased board independence were associated with high quality financial reporting through the deterioration of earnings management behaviour. In another study by Fodio, Ibikunle and Oba (2013), reported that board composition was negatively and significantly associated with earnings management. Arabborzoo, Rashidpuran and Arabi (2015), found that outside board members did not significantly affect earnings quality of firms listed in Tehran Stock Exchange. In Sri Lanka, Kankanamage (2015) adopted ordinary least squares regression and revealed that there was a significant relationship between board composition and earnings management of the firms. A study by Yohan (2016) in Korea found that outside directors tended to have a positive effect on firm value

but not on earnings quality.

Based on the above arguments, the study therefore, hypothesized that: Ho1: Board diversity has no significant effect on earnings management of listed Deposit Money Banks in Nigeria.

2.3.6 Audit Committee and Earnings Management

29

A study by Chandrasegaram, Rahimansa, Rahman, Abdullah and Nik Mat (2013) found that audit committee played a significant role in mitigating earnings management of firms. Also, Yasser and Al Mamun (2016) found audit committee size was positively associated with financial reporting quality. The results indicated that the audit committee was a less significant factor in corporate governance than suggested by many previous researchers and policy makers. Ayemere and Elijah (2015) found that, audit committee characteristics had a constraining effect on earnings management. Another study by Xi'an and Xi'an (2012) found that participation of women in audit committees had a negative and significant influence on earnings management of firms. On the other hand, Ioualalen, Khemakhem and Fontaine (2015) found that audit committee diversity did not have any significant impact on earnings management of selected Canadian firms. Following the

positions established in previous researches, the study therefore, hypothesized as follows.

Ho2: Audit committee has no significant moderating effect on the relationship between women director, board ownership and earnings management of listed Deposit Money

Banks in Nigeria.

2.4 Theoretical Framework This section relates each of the theory adopted with the variables used in the study. Corporate monitoring by diverse board in terms of women director and foreign director is expected to constrain managers’ behavior because corporate monitoring by diverse board may force managers to focus more on corporate performance and less on opportunistic or self-serving behaviour. If corporate board diversity enhances monitoring, it is expected to be associated with lower use of discretionary loan loss provision in the banks. Thus, organizations with more diverse senior managers are expected to reduce the level of earnings management.

The Agency theory view board of directors as the agent of the shareholders and as such there is need for it (board of directors) to act in the best interest of the shareholders. In this situation, sometimes the agent may not act in the best interest of the shareholders which often results in an agency loss situation. Advocates of the agency approach view the manager (directors) as an economic institution that will mitigate problems and serve as guardian to shareholders (Hermalin & Weisbach 2003). Based on the relevance and usefulness of agency theory in resolving conflict between principal (owners) and agent (managers) of organization. The study therefore adopted agency theory to underpin the study.

3. Methodology The study adopts the ex-post facto research design because it conducted based on positivism paradigm and quantitative approach. The data were sourced from the annual reports and accounts of Fifteen (15) Deposit Money Banks listed on the NSE at 31st December, 2008 and remained listed up till 2015. Multiple regression technique was adopted and the analysis was carried out using Stata 13 as the tool of data analysis. The first regression residual was used to determine the Earnings Management (here after refer to as EM) of the banks and in the sample. In the second regression the effects of board diversity and audit committee on earnings management of the banks was

measured.

30

The residual of the following model is used to estimate EM of the banks. The model adopted was Chang, Shen and Fang (2008) which was specifically built for financial sector and it was through this model that the residual were arrived at for the second model. DLLPit /TAt-1 = LLPit/TAt-1 – {α0 1/TAt-1 + α1 LCOit/TAt-1 + α2 BBALit/TAt-1}……………… (i)

DLLP represents Discretionary loan loss provision; LLP = Loan loss provision; LCO = Loan Charge-off; BBAL = Beginning Balance of loan loss; TAt-1 = Lagged Total Assets; α0 = Constant. The second model is presented in the following equation which is used to estimate the measure the predictive ability of board diversity and audit committee on EM. EMit = β0t + β1Wdirit + β2Bownit + β3Fdirit + β4Bsizeit + β5Bodcit + β6Accit + µit …….. (ii) EMit = β0t + β1Wdirit + β2Bownit + β3Fdirit + β4Bsizeit + β5Bodcit + β6Wdirit*Accit + β7Bownit*Accit + µit ……………………………………………………………………….....… (iii) Wdirit is Women director measured as number of women on board over the total number of board members. (Bathula, 2008); Bownit is Board Ownership measured as the percentage of share held by the board of directors (Farouk, 2014); Fdirit is Foreign Director represented the number of foreign director divided by total number of board members (Abdul Rauf, Johari, Buniamin, & Abd Rahman, 2012); Bsizeit is Board Size measured as the number of Board members in a particular year. (Daghsnii, Zouhayer & Mbarek, 2016); Bodcit is Board Composition measured as the ratio of Non-executive director to total number of directors. (Arabborzoo, Rashidpuran & Arabi, 2015); and Accit

is a composite index representing Audit Committee which is made of: (i) Audit committee size, the number of committee members, (ii) Audit committee composition, the number of shareholders over the total number of committee members, (iii) Audit committee meeting, the number of times the committee held meetings in a year. β1 - β7 = Coefficient of explanatory variables, βo = Constant or Intercept and µ = Error Term. Robustness tests such as multicollinearity, heteroscedasticity, normality test of the standard error and

langrage multiplier tests were conducted to validate the results.

4.1 Results and Discussion This section first established the usefulness and the strength of Chang, Shen and Fang (2008) model in Nigerian Banking Sector. It further present, interprets, analyzes and discusses the regression results. Based on the F-ratio and the probability value, the first hypothesis of the study was tested, while the R square of the two regression (Unmoderated and Moderated) were compared and use as a basis of testing the seconf

hypothesis of the study.

4.1.1: Summary of Descriptive Statistics I The minimum value for discretionary loan loss provision in the banks stood at -7.44e-09 within the study period, while the maximum value stood at 3.45e-09. However, on average, the earnings management of banks was -1.40e09. This indicates a decreasing level of manipulation for discretionary loans. The ratio of loan loss provision has their minimum and maximum value stood at -6.94e-09 and 2.50e-09 respectively. The mean of loan loss provision was -1.20e-10 implying that the ratio of loan loss provision to total assets is decreasingly managed. The inverse of total assets value for both minimum and maximum, stood at 2.30e-10 and 5.45e-09. The mean of the inverse of total assets for all the banks was 1.45e-09. The minimum value for loan charged-off for banks within the

31

study period stood at -2.60e-10, while the maximum value is 2.86e-09. The mean value reported for loan charged-off was 8.91e-11 for banks. The beginning balance of loan loss provision has a minimum value of -8.96e-09 and a maximum value of 1.49e-09. However, the mean value for beginning balance of loan loss provision was 1.41e-10. Below is the inferential statistics.

Table 1: Regression Results of Chang, Shen and Fang (2008) Model Variables Coeff T-Stat Prob. Constant 9.80e-11 1.37 0.173 LLPTA 1.167 16.2 0.000 TA -0.98 -23.6 0.000 LCOTA 2.337 15.7 0.000 BBALTA 1.041 17.8 0.000 R2 0.9178 Adjusted R2 0.9150 F-Statistics 321.07 F-Significance 0.0000

Source: Result output from Stata 13

From the cummulative result, the model records an R2 of 0.9178 which showed the extent to which the discretionary loan loss provision was explained by loan loss provision scaled by total assets, inverse of total assets, loan charge off sacled by total assets and beginning balance of loan loss provision scaled by total assets. This showed the strength of the model as developed by Chang et al (2008) and its applicability as well as the usefulness of the model in ascertaining the extent of earnings management of listed Deposit Money Banks in Nigeria. Hence, this justified the adoption of the model in this study. Furthermore, the Fisher exact test (F-Statistics) value of 321.07 indicated that the model of the study was well fitted and as such the variables in the model were properly

selected, combined and used.

All the independent variables (Loan loss provision scaled by total assets, Inverse of total assets, loan charge off to total assets, Beginning balance of loan loss provision to total assets) in the model were all found to have significant effect on discretionary loan loss provision. The signs accompanying each of the individual coefficient estimate was as

specified in the model which further substantiates the appropriateness of the model.

4.1.2: Summary of Descriptive Statistics II The summary of statistics result indicates that the minimum value for earnings management in banks was 0.00003, while the maximum value was 0.14507. This indicates that the level of earnings manipulations of banks in Nigeria is minimal. To further substantiate this, the mean value of 0.007 indicates that the earnings manipulation level of banks was far below 1%. This could be associated with the adoption of both corporate governance and international financial reporting standard for best practices. Women director representation on board was zero percent some banks, while the highest representation stood at 60%, however, on average every banks have at least 14% of women on board. The minimum value for board ownership stood at 1.02%, while the maximum value is 55.19%. On average the total percentage of shares held by banks stood at about 9%. For foreign directors’ representation on board, the minimum was

32

zero, while the maximum was 50%. On average, the banks have at least 10% foreign directors on board. The minimum and maximum number of board members stood at eight and twenty one for banks within the study period, while on average, each bank is deemed to have at least about 15 numbers of board members. The composition of the board in terms of non-executive directors to total number of directors is reported to have 21% on minimum and 88% as maximum for banks, while on the average, the non-executive directors occupies about 57% of the board members in banks. The inferential statistics is presented below.

Table 2: Summary of Random Effect Model Unmoderated Moderated

Variables Coeffi Z-Stat Prob Coeffi Z-Stat Prob Constant 0.186 6.28 0.000 0.191 3.32 0.001 Wdir -0.011 -1.71 0.086 -0.054 -1.43 0.154 Bown 0.019 4.12 0.000 0.018 0.92 0.357 Fdir 0.030 4.79 0.000 0.026 3.85 0.000 Bsize -0.001 -1.63 0.104 -0.001 -1.43 0.154 Bodc -0.044 -2.88 0.004 -0.037 -2.39 0.017 Acc -0.015 -5.15 0.000 -0.016 -2.81 0.005 Wdirac 0.004 1.16 0.245 Bownac -0.001 -0.05 0.963 R2 Within 0.4593 0.4588 R2 Between 0.2882 0.3069 R2 Overall 0.3521 0.3701 Wald Chi2 88.57 83.66 Prob. 0.0000 0.0000

Source: Result output from STATA 13

The results of the robustness tests include: heteroscedascticity test (p-value = 0.01) indicated the unfitness of the OLS and we proceeded with the GLS; the hausman specification tests for unmoderated and moderated model (p-value = 0.9385 and 0.9303) respectively, which indicated appropriateness of the random effect for the panel regression; Langrange Multiplier tests for unmoderated and moderated model (P-value = 0.01 and 0.01) respectively, which indicated presence of panel effect and hence the random effect used for this study. The regression results follow:

The significance influence of the audit committee variable on earnings management of Banks in the first model (unmoderated) showed that the variable can be used as a moderator, however, on comparison of the overall R2 between the unmoderated (0.3521) and moderated (0.3701) models, it indicated that the interaction between women directors and audit committee, board ownership and audit committee accounted for significantly more variance than when they were used as single variables. This implied that, the moderated model explained the behavior of earnings management better in relation to the independent variables of the study when compared with the unmoderated model. On the other hand, the test for significance difference which recorded a chi-square value of 34.20 and p-value of 0.0000 (see Appendix) indicated that there was a significant differences between the unmoderated and moderated models.

33

Under the unmoderated model, five of the variables (women directors, board ownership, foreign director, board composition, audit committee) were significant in influencing earnings management except for board size. Women director, board composition and audit committee were negatively related to earnings management under the unmoderated model which is also in line with the studies conducted by Lakhal, Aguir, Lakhal and Malek, (2015), Yohan (2016) and Xi'an and Xi'an (2012), but contrary to those of Hussaini and Gugong (2015), Arabborzoo, Rashidpuran and Arabi (2015) and Yasser and Al Mamun (2016). Also, board ownership and foreign director were positively related to earnings management of banks which also concur with the findings of Hooghiemstra, Hermes, Oxelheim and Randoy (2015) and Ogbonnaya, Ekwe and Ihendinihu (2016), but disagree with the findings of Ramadan (2015) and Abdul Rauf, Johari, Buniamin and Abd

Rahman (2012).

For the moderated model, only three of the variables (foreign directors, board composition and audit committee) were having significant effect on earnings management of banks except for women directors, board ownership and board size. Among the significant variables, board composition and audit committee were found to have negative relationship with level of earnings management of banks while foreign

director has positive relationship with the level of earnings management.

The two moderated variables were found to have insignificant effect on earnings management after the moderation, implying that the variables only affect earnings management individually than when they are moderated with audit committee. Another interesting findings this moderation is that the direction of the relationship changes. Women directors changed from negative to positive after being moderated with audit committee, while board ownership changed from positive to negative after being moderated with audit committee which may have resulted in its insignificant influence on

earnings management.

Based on the forgoing analysis in respect of all the variables, the hypothesis one of the study which stated that board diversity has no significant effect on earnings management of listed Deposit Money Banks in Nigeria was tested. With respect to the findings showing that Wald Chi2 of 88.57 which was significant at the level of 1%. This implied that the independent variables significantly predicted the outcome of earnings management in listed deposit money banks in Nigeria. This therefore, provided enough evidence to reject

null hypothesis I of the study.

Based on the findings and analysis in respect to moderated model of the study. The hypothesis which was formulated earlier in section one stating that audit committee had no significant moderating effect on the relationship between women directors, board ownership and earnings management of listed Deposit Money Banks in Nigeria was tested using the R2 statistics. Based on the R2 overall in respect of unmoderated and moderated models, they both record 0.3521 and 0.3701 respectively. This implied that the high R2 recorded for moderated model as against the unmoderated model was an indication that the audit committee significantly moderated the relationship between women director, board ownership and earnings management. Therefore, this provided sufficient evidence of rejecting null hypothesis two of the study.

5.1 Conclusion and Recommendations

34

From the findings, the study concluded that board diversity and audit committee gender significantly affected the extent of earnings management in listed Deposit Money Banks in Nigeria. While audit committee can be concluded to have played a significant moderating role between the two moderated variables. Moderated board ownership is better than moderated women director because of the changes in direction of effect on earnings management. Overall, when diversity variables were improved upon, alongside increase in audit committee variables, the earnings quality of banks in Nigeria was guaranteed.

It is therefore recommended that management of the banks should increase the audit committee variables as they help checkmate the activities of the management and reduce the level of earnings management. Also, the management should consider members of the board who have shares when constituting the audit committee. In addition, women directors who partake in audit committee should be carefully selected by the management

to ensure their full participation.

References

Abdul Rauf, F.H., Johari, N.H., Buniamin, S. and Abd Rahman, N.R. (2012). The Impact of Company and Board Characteristics on Earnings Management: Evidence from Malaysia. Global Review of Accounting and Finance, 3(2), 114-127

Arabborzoo, A. Rashidpuran, S. Arabi, A. (2015). The impact of corporate governance on earnings quality. Journal of Scientific Research and Development, 2(1), 127-132.

Arun, T.G., Almahrog, Y.E. and Aribi, Z.A. (2015). Female Directors and Earnings Management: Evidence from UK Companies. International Review of Financial Analysis. 39(C), 137-146

Ayemere, I.L. and Elijah, A. (2015). Audit Committee Attributes and Earnings Management: Evidence from Nigeria. International Journal of Business and Social Research, 5(4), 14-23

Baccouche, S Hadriche, M and Omri, A. (2014). Multiple Directorships and Board Meeting frequency: Evidence from France. Applied Financial Economics, 24(14), 983-992.

Barnea, A., Ronen, J. and Sadan, S. (1976). Classificatory smoothing of income with extraordinary items. The Accounting Review, 110-122.

Bathula, H. (2008). Board Characteristics and Firm Performance: Evidence from New Zealand. An Unpublished thesis submitted to Auckland University of Technology in fulfilment of the requirements for the degree of Doctor of Philosophy.

Bello, A. (2005). Economics of Financial Reporting: A test on Nigerian Banks. An unpublished Seminar paper presented at the department of Accounting, Ahmadu Bello University, Zaria.

Burton, C. & Ryall, C. (1995). Managing diversity. In Enterprising nation: reviewing Pacific Century, Research Report Australian Government Publishing Service, Canberra, 2,765-814.

CBN, (2006). Code of Corporate Governance for Banks in Nigeria post consolidation. Chandrasegaram, R., Rahimansa, M.R., Rahman, S.K.A., Abdullah, S. and Nik Mat, N.

(2013). Impact of Audit Committee Characteristics on Earnings

35

Management in Malaysian Public Listed Companies. International Journal of Finance and Accounting, 2(2), 114-119.

Chang, R. D., Shen, W.H.A. and Fang, C. J. (2008). Discretionary Loan Loss Provisions and Earnings Management for the Banking Industry. International Business & Economics Research Journal, 7(3), 9-20.

Copeland, R. M. (1968). Income smoothing. Journal of Accounting Research, 6, 101-116. Daghsnii, O; Zouhayer, M. and Mbarek, K.B. (2016). Earnings Management and Board

Characteristics: Evidence from French Listed Firms. Account and Financial Management Journal, 1(2), 92-110

Dimitropoulos, P. (2011). Corporate Governance and Earnings Management in the European Football Industry. Journal of European Sport Management Quarterly, 11(5), 495-523

Farouk M. A. (2014). Possession Formation and Earnings Management of Listed Chemical and Paints Firms in Nigeria. Being an M.Sc Thesis submitted to the school of Postgraduate Studies, Ahmadu Bello University, Zaria.

Fei, C.C. (2015). The Impact of Managerial Opportunism on Earnings Reliability. International Journal of Economics and Finance, 7(10), 222-234

Fodio, M. I; Ibikunle, J; and Oba, V.C. (2013). Corporate Governance Mechanisms and Reported Earnings Quality in Listed Nigerian Insurance Firms. International Journal of Finance and Accounting. 2(5), 279-286

Gulzar, M. A. and Wang, Z. (2011). Corporate governance Characteristics and Earnings Management: Empirical Evidence from Chinese Listed Firms. International Journal of Accounting and Financial Reporting, 1(1), 133- 151.

Guo, J., Huang, P., Zhang, Y. and Zhou, N. (2014). Foreign Ownership and Real Earnings Management: Evidence from Japan. Being a Paper presented at the 2012 American Accounting Association (AAA) Annual Meeting and the 2012 AAA Mid-Atlantic Meeting.

Hermalin, B. E. and Weisbach, M. S. (2003) Boards of directors as an endogenously determined institution: A survey of the economic evidence. Economic Policy Review, 9, 7-26.

Hooghiemstra, R., Hermes, C., Oxelheim, L., & Randoy, T. (2015). The impact of board internationalization on earnings management. SOM Research Reports, 15010 (I&O), University of Groningen.

Holtz, L. and Neto, A.S. (2013). Effects of Board of Directors’ Characteristics on the Quality of Accounting Information in Brazil. Paper presented at the VII Anpcont Congress, Fortaleza, CE, Brazil, 25(66), 255-266

Hussaini, B. and Gugong, B.K. (2015). Board Characteristics and Earnings Management of Listed Food and Beverages Firms in Nigeria. European Journal of Accounting, Auditing and Finance Research, 3(8), 25-41.

Ibrahim, I. (2015). Board Characteristics and Earnings Management of Listed Foods and Beverages Firms in Nigeria. Being a M.Sc Dissertation Submitted to the School of Postgraduate Studies, Ahmadu Bello University, Zaria

Ishaq, M.T. (2017). Effect of Board Diversity on Earnings management of listed Conglomerates Firms in Nigeria. Being an M.Sc Thesis submitted to the School of Postgraduate Studies, Ahmadu Bello University, Zaria.

Ioualalen, L., Khemakhem, H. and Fontaine, R. (2015). The Impact of Audit CommitteeCharacteristics on Earnings Management: A Canadian Case Study. Case Studies in Business and Management, 2(1), 78-96.

36

Kankanamage, C. A. (2015). The Relationship between Board Characteristics and Earnings Management: Evidence from Sri Lankan Listed Companies. Kelaniya Journal of Management, 4 (2), 36-43

Kantudu, A.S and Samaila, I.A. (2015). Board Characteristics, Independence Audit Committee and Financial Reporting Quality of Oil Marketing Firms: Evidence from Nigeria. Journal of Finance, Accounting and Management, 6(2), 34-50.

Kurawa, J. M. and Saheed, A. (2014). Corporate governance and earnings management: An Empirical Analysis of firms in petroleum and petroleum products distribution in Nigeria. Research Journal of Accounting, 2(2), 10-14

Lakhal, F., Aguir, A., Lakhal, N., and Malek, A. (2015). Do Women On Boards And In Top Management Reduce Earnings Management? Evidence in France. The Journal of Applied Business Research, 31(3), 1107-1118.

Lei, K. (2003). Earnings management and corporate governance in UK: the role of board of directors and audit committee. Journal of Accounting and Finance, 5(2), 23-39.

Marimuthu, T. (2008). The role of Private Sector in Higher education in Malaysia. in D. Johnson & Maclean, R. (Eds.), Teaching: Professionalization, Development and Leadership, 271-282.

Mazur, B. (2010). Cultural Diversity in Organisational Theory and Practice. Journal of Intercultural Management. 2(2), 5–15

Mustafa; S; Mehmet, A; and Suleyman, I.C. (2014). The Impact of Corporate Ownership Structure and Board Size on Earnings Management for a Sample of Turkish Firms. International Journal of Business & Management, 9(12), 123-138.

Oba, V.C., (2014). Board Dynamics and Financial Reporting Quality in Nigeria. Review of International Comparative Management, 15(2): 226-237.

Ogbonnaya, A.K., Ekwe, M.C. and Ihendinihu, J.U. (2016). Effect of Corporate Governance and Ownership Structure on Earnings Management of Brewery Industry. European Journal of Accounting, Auditing and Finance Research, 4(7), 35-45

Omoye, A.S and Eriki, P.O. (2014). Corporate Goverance Determinants of Earnings Management: Evidence from Nigerian Quoted Companies. Mediterranean Journal of Social Sciences, 5(23), 553-564

Patrick, E.A; Paulinus, E.C. and Nympha, A.N. (2015). The Influence of Corporate Governance on Earnings Management Practices: A Study of Some Selected Quoted Companies in Nigeria. American Journal of Economics, Finance and Management, 1(5), 482-493

Qi, J. and Tian, S. (2012). Female Board Presence and the likelihood of Financial Restatement. American Accounting Association, 26(4), 607-629.

Ramadan, I.Z. (2015). Does Ownership Structure Affect Jordanian Companies’ Tendency to Practice Earnings Management? Asian Journal of Finance & Accounting, 7(2), 281-291.

Razek, M. (2012). Board Characteristics and Earnings management in Manufacturing Companies in Nigeria. Research Journal of Finance and Accounting, 3(8), 15-31.

Rose, C. (2007). Does female board representation influence firm performance? The Danish evidence. Corporate Governance: An International Review, 15 (2), 404-13.

37

Saleh, M.N. and Haat, I.T. (2014). Earnings Management and Board Characteristics: Evidence from Malaysia. Journal Pengurusan, 24, 77-103

Salihi, A.A. and Jibril, R.S. (2015). The Effect of Board the Size and Audit Committee the Size on Earnings Management in Nigerian Consumer Industries Companies. International Journal for Innovative Research and Development, 4(3), 84-91

Shehu, U. H. (2012). Firm Characteristics and Financial Reporting Quality of Quoted Manufacturing Firms in Nigeria. An unpublished dissertation submitted to the school of postgraduate studies, Ahmadu Bello University, Zaria.

Shehu, U. H., & Farouk, M. A. (2014). Board of Director's characteristics and Performance of listed Deposit Money Banks in Nigeria. Journal of Finance and Bank Management, 2(1), 89-105.

Shehu U.H and Ibrahim, G. (2014). Governance Attributes and Real Activities Manipulation of Listed Manufacturing Firms in Nigeria. International Journal of Accounting and Taxation, 2(1), 37– 62

Susanto, Y.K. (2016). The Effect of Audit Committees and Corporate Governance on Earnings Management: Evidence from Indonesia Manufacturing Industry. International Journal of Business, Economics and Law, 10(1), 32-37.

Vafeas, N. and Theodorou, E. (1998). The relationship between board structure and firm performance in the UK. British Accounting Review, 30, 383-407.

Xi'an, B.Q. and Xi'an, G.T. (2012). The Impact of Audit Committees Personal Characteristics on Earnings Management. The Journal of Applied Business Research, 28(6), 1331-1343

Yasser, Q.R. and Al Mamun, A. (2016). Audit committee structure and earnings management in Asia Pacific. Economics and Business Review, 2(16), 66–84

Yohan, A. (2016). Effect of Outside Director’s Quality on Firm Value and Earnings Quality. Research Journal of Business Management, 11, 39-45

Yu, M. (2013). State ownership and firm performance: Empirical evidence from Chinese listed companies. China Journal of Accounting Research, 6, 75–87

Zhang, Y. and Li, Y. (2007). The Effect of the Directorate Characteristics on Earnings Management Based on Chinese Market Data. Available at http://ieeexplore.ieee.org/document/

Zhang, Y. & Uchida, K., (2011). Corporate governance, investor protection, and earnings management: New international evidence, 6–19.

38

Public Spending, Fiscal Sustainability and Macroeconomic Performance in

Nigeria

By Olufemi Muibi Sabiu

1 Introduction

Fiscal sustainability has drawn so much attention, especially in developing countries, in terms of debt profiles and their impact on their growth trajectories. Discussions on fiscal sustainability have featured prominently among policy makers and mainstream economists in developing countries. This increased attention on fiscal sustainability is attributed to the fact that most developing countries often experience significant upswings and downswings in their revenue generation and thus leading to significant

fiscal constraints in their development plans.

Nigeria’s fiscal history presents a typical of this trajectory. Nigerians fiscal spending is substantial dependent on oil revenue. More than 90% of Nigerian government budgetary finances is sourced from revenue generated from oil. The fluctuations in oil prices therefore translate directly to fluctuation in government revenue. The fall in oil prices has consistently led to increased borrowing in financing the budget since 1980s when oil price experienced significant downward swing. Nigeria’s total debt profile maintained a consistent increase, rising from per cent of gross domestic product in 1981 to 39.1in

1989. It further increased to a peak of 56.6 per cent in 1990, (CBN, 2016).

Following the debt cancellation package which the country negotiated and obtained from the Paris Clubs, there was a gradual reduction from the 56.6 % 1.5 percent debt/GDP in 2006. For the period 2006 to 2009 and remained at single digit till 20101 and marginally increased to an average of 10 per cent between 2011 and 2013. Following the swings in international oil prices, in 2014, the debt profile upward swing. It rose to 21.38 per cent in 2016 from 15.86 per cent and 14.20 per cent in 2015 and 2014 respectively though the debt rate is still less than the 56 per cent global benchmark for developing countries like Nigeria, the 2016 figure is higher than the DMO’s benchmark of 20% and its prediction

for 2016 of 13.5 per cent (DMO, 2016).

The preceding evidence gives much impetus for a more systematic study analysis of the sustainability of the fiscal policy and its effect on macroeconomic performance in Nigeria. Unlike previous studies that have traditional focused on the debt-GDP growth nexus only, this study push further to link the sustainability of government fiscal policies to macroeconomic economic performance, using a more robust construct of macroeconomic

performance in Nigeria.

The paper is further motivated by the need to analyze the fiscal position of the country and to better understand the risk the country is facing, especially within the present economic context. Notwithstanding the relatively low debt to GDP ratio of the country in recent years, as debt continue to rise to historical levels, the risk premia begins to rise

39

sharply, and so the economic situation may begin to deteriorate. Indeed, the debt servicing/revenue ratio which stood at 64 percent in 2016 compared with 32.7 and 29 percent for 2015 and 2014 respectively has exceeded the benchmark of 28 percent by

DMO indicating a more vulnerable situation of the fiscal sustainability in the future.

Consequently, the current and future economic growth performance can be significant constrained as government may not have much resources to finance development projects. Finally, a fiscally sound economy will be able to finance public investment and

at the same time attract foreign investment through its fiscal credibility.

The paper extends and contributes to the extant literature in two ways. It adopts a numerical expenditure rule as alternative to the conventional non numerical approached among previous studies on debt/growth nexus in Nigeria, (see Omotosho, Bawa and Dagura 2016, for instance). This empirical strategy follows the procedure employed in past studies (such as Baharumshah and Lau, 2007 and Miyazaki, 2011) that have dealt with the sustainability problem in other countries by considering structural change. The method not only established fiscal sustainability but also provide a general framework for analysis the dynamics of the fiscal sustainability intertemporally over a given period. Secondly, the paper experiments with deriving a unique index for macroeconomic performance as against the use of economic growth common among previous studies. The new index provides opportunity to capture all aspects of economic performance indicators.

The remainder of the paper is structured as follows: sections 2 outlines the theoretical and empirical issues in fiscal sustainability. Section 3 presents the empirical methodology while section 4 discusses the empirical results and the concluding remarks are covered in

the section 5

2. Theoretical and empirical issues in fiscal sustainability

The Treasury (2013) defines fiscal sustainability in relation to the affordability of government taxation and spending programmes. In simple terms, fiscal just refers to government spending and investing activities and how these are financed through taxes, debt and other liabilities. Sustainability means having the ability to maintain or support government programmes in the future. So, fiscal sustainability refers to whether the Government can maintain current policies without major adjustments in the future. Fiscal sustainability represents broadly the ability of government to sustain its current spending, tax and other policies in the long run without threatening government solvency or defaulting on some of its liabilities or promised expenditures. Tanner (2013) argues that for a government to meets all its present and future obligation, its streams of revenues must at least be enough over a period to offset and services its obligation to service its obligations to preclude either default or restructuring. Several studies have rendered theoretical frameworks and empirical evidence on fiscal sustainability. Tanzi (2011) states that public debt is an important fiscal policy tool and sustaining a balance between revenue inflow and spending is key in fiscal management. According to the paper, the idea of tax smoothing, which is essential to provide a steady flow of public goods and services and run a countercyclical fiscal policy, depends on the capacity of government to borrow during recessions and repaying debt during booms. In this sense, “some economists have advocated borrowing to finance public investment, in

40

which has been referred to as the “golden rule”. This argument fundamentally relies on two basic assumptions: - first; “because public investment creates assets that favor future generations, thus the latter should pay for it and second; that public investment is always productive” (Tanzi, 2011). Even though debt can be employed as a tool to bridge fiscal constrain, thus improving macroeconomic performance and promoting welfare, excessive debt accumulation can lead to a fiscally unsustainable situation, with severe negative effects on macroeconomic stability and economic growth performance. Reinhart and Rogoff (2011) documented large number of countries with historical “surges in public debt” episodes of excessive debt accumulation. The conclusion from this paper and many other subsequent studies that examined the issues was that countries faced chronic fiscal problems mostly because governments systematically overspend and do not have the political will or ability to tax effectively and efficiently or a combination of the two. In this regard it is argued that significant asymmetry between expenditure and taxes has led to fiscal illusion when the public sector relies heavily on borrowing to finance fiscal deficits. A fiscal illusion arises when debt finance is substituted for tax finance thereby causing people to underestimate the real price of public goods and thereby increase their demand for more spending” (see Aronson and Ott, 1996) In view of the large fiscal deficits and debt overhang that characterized most of the developing countries and the resultant retarded growth experienced by majority of the economies in Africa Studies have also examined the sustainability of fiscal policy and debt profile of developing countries. Most of these studies fundamentally explain the nexus between debt and growth, as well as setting a debt threshold upon which debt becomes inefficient for driving growth (Schneider (2006); Redzepagic (2008); Westerlund and Prohl, 2010; Omotosho, Bawa and Doguwa, 2016). Other studies have focused on the relationship between fiscal rules, sustainability of fiscal policy and the outcome of economic performance (Fat´as ,2010; Franco and Zotteri, 2010; and Rose ,2010), the relationship between fiscal reform and fiscal sustainability and emphasize the role of fiscal rules in ensuring the sustainability of fiscal policy, (Franco and Zotteri, 2010). Some other studies have also extended the analysis to the sustainability of fiscal policy in general, (Bohn, 2008; Afonso and Rault, 2009; Legrenzi and Milas, 2010; Abdullah, Mustafa and Dahalan (2012). The lack of consensus and a common benchmark for determining sustainable fiscal spending and borrowing, Burnside (2012) provided a new framework for assessing key indicators of the health and soundness of fiscal policy stands. This new approach which is adopted in this paper provide an alternative strategy to test for the robustness of existing framework and determine if the methodology approached adopted constitute a major factor in the divergent views expressed in the literature with respect to fiscal sustainability in terms of ; estimation of government’s ability to borrow as well as finance its debt servicing and repayment; prediction of the onset of fiscal crises that may be lurking; assessment of the fiscal risk associated with contingent liabilities; and the assessment of prior fiscal policy record and discussion of future policy choices.

41

3. Analytical Framework

3.1 Theoretical Model for Fiscal Sustainability Test

Following the theoretical argument of Bohn (2007) as adopted by ( Silvestrini 2007; Deyshappriya, 2012; and Muzenda, 2014), the Nigeria’s government fiscal stance can be construed in term of intertemporal budget constrain (IBC) which defines the extent to which government has exceeded it budget constrain and how this deficits can be financed. Thus, the nominal IBC can be written as;

∆𝐷𝑡 = 𝐺𝑡 + 𝑟𝑡𝐷𝑡−1 − 𝑅𝑡 (1)

In equation 1 above 𝐷𝑡−1 is government total debt stock and thus ∆𝐷𝑡 denotes the change

in stock of debt while 𝑟𝑡𝐷𝑡−1 is the interest payment of debt financing; 𝑅𝑡 is government

revenue while 𝐺𝑡 is government total expenditure and 𝑟𝑡 is the prevailing real interest

rate. The right hand side of equation 1 represents the primary balance which can be expressed as n term of government spending to have:

𝐺𝑡∗ + (1 + 𝑟𝑡)𝐷𝑡−1 = 𝑅𝑡 + 𝐷𝑡 (2)

The debt profile can also be seen from future streams of income payment to be made if the debt is to be liquidated. Thus, the current debt stock could be expressed as equivalent of the discounted value of future streams of income that accrues from fiscal surpluses (𝑆𝑡+1), calculated as 𝑆𝑡+1 = 𝑅𝑡+1 − 𝐷𝑡+1 , for a period j. Similarly, if instead of surplus, the

fiscal balance is in deficit and hence debt accumulates further. The net current debt stock will therefore be cumulative sum of the discounted surpluses and deficits of the given period. Hence, expressing equation 2 in term of present and future discounted fiscal

surplus and debt values such that the debt yield for subsequent periods gives;

𝐷𝑡 =1

1 + 𝑟𝑆𝑡+𝑗+1 +

1

1 + 𝑟𝐷𝑡+𝑗+1, 3

Equation 3 specification builds on the assumption that (i) time is discrete, (ii) debt has a maturity of one period, (iii) debt is real (in other words, its face value is indexed to the

prevailing price level) and (iv) debt issued at date 𝑡 − 1 pays a real interest rate 𝑟𝑡−1. (𝑣)𝑟𝑡 is assumed to be stationary around its mean 𝑟.

Following Miyazaki 2011, by a simple iterative substitution, and taking conditional expectation, equation 3 was solved recursively such that equation 3, becomes;

𝐷𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

𝐸[𝑆𝑡+𝑗+1] + lim𝑗→∞

1

(1 + 𝑟)𝑗+1𝐸[𝐷𝑡+𝑗+1] (4)

As 𝑗 → ∞ the present value of expected debt to GDP ratio should converge to zero such

that

lim𝑗→∞

1

(1 + 𝑟)𝑗+1𝐸[𝐷𝑡+𝑗+1] = 0 (5)

42

and the necessary condition for fiscal sustainability is derived by setting the current value of the outstanding government debt (𝐷𝑡) 𝑡𝑜 be equal to the expected present value of

future budget surplus

𝐷𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

𝐸[𝑆𝑡+𝑗+1], (6)

Equations 5 & 6 guarantees that the intertemporal budget balance (IBB) holds and the

government solvency constrain and transversality condition is also satisfied;

Equation 5 can also be used to articulate two cases of fiscal sustainability analysis.

(i) If lim𝑗→∞

1

(1+𝑟)𝑗+1 𝐸[𝐷𝑡+𝑗+1] < 0, then expected discounted future primary surpluses

exceed the present value of public debt suggesting that over time government accumulates a net tax revenue.

(ii) If lim𝑗→∞

1

(1+𝑟)𝑗+1 𝐸[𝐷𝑡+𝑗+1] > 0, then the present value of government debt

exceeds the expected primary surpluses and the government is under pressure to borrow persistently overtime to cover the shortfalls and hence accumulate debt to meet its interest obligation on past debt incurred.

The optimal condition is when equation 5 holds which suggests that debt is solvent when the transversality condition ensures the non-explosiveness of public debt condition (and when there No Ponzi Game, hereafter, NPG condition) is fulfilled. That is (i) the present value of all future debt balances must be zero (ii) the current debt is offset by the sum of current and expected future discount surpluses and (iii) the budget constrain holds in

present value terms as expressed in equation 6.

3.2 Empirical Model for Testing Fiscal Sustainability

Equation 4 can therefore be used to derive a baseline model for testing the fiscal sustainability condition. Since the study is time series, then equation 4 needs to be converted to a testable empirical equation by taking its first difference to yield:

∆𝐷𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

[∆𝑆𝑡+𝑗+1] + lim𝑗→∞

1

(1 + 𝑟)𝑗+1[∆𝐷𝑡+𝑗+1] (7)

From equation 1, ∆𝐷𝑡 = 𝐷𝑡 − 𝐷𝑡−1, then equation 7 can be rewritten as:

𝐺𝑡 + 𝑟𝑡𝐷𝑡−1 − 𝑅𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

[∆𝑆𝑡+𝑗+1] + lim𝑗→∞

1

(1 + 𝑟)𝑗+1[∆𝐷𝑡+𝑗+1]

Based on equation 5 & 6 and with the assumption that intertemporal budget balance

condition holds, then equation 7 can be rewritten as;

𝐺𝑡 + 𝑟𝑡𝐷𝑡−1 − 𝑅𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

[∆𝑆𝑡+𝑗+1] (8)

43

Equation 9 fundamentally, implies that the sum of the present value of discounted current and future budget surplus will be equal to the amount needed to just repay the principal amount of debt and the interest yield on the debt. Thus, equation 9 guarantees that IBC or NPG is attained and gives the condition to be met for the current expected path of government fiscal activity to be sustainable in the long run (Mahmood and Rauf, 2012)

The Equation 9 thus forms the basis for the empirical test for fiscal sustainability for existence of the IBC and NPG. The common approach of testing this sustainability condition is to test for the stationarity of the terms in the left hand side of equation 9 by imposing the cointegrating vector (1,1,-1) (Miyazaki 2011). In this sense, the stationarity

test is explicitly carried out on the discounted debt series 𝐷𝑡 and to test for the

cointegration between the government expenditure 𝐺𝑡 and government revenue 𝑅𝑡. For

empirical purpose the left-hand side of equation 8 can be subsumed into a single variable thus; 𝑋𝑡 = 𝐺𝑡 + 𝑟𝑡𝐷𝑡−1, and the right-hand side be subsumed as streams of income from

revenue such that equation 9 becomes:

𝑄𝑡 = ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

[∆𝑆𝑡+𝑗+1] (9)

The LHS of the equation 9 expresses the budget balance captured by 𝑄𝑡 (that is 𝑄𝑡 =𝑅𝑡 − 𝑋𝑡) while the RHS is the discounted debt stock. To explicitly test for the cointegration

in the regression equation 9 is simplified as follows:

𝑅𝑡 = 𝛿 + 𝜑𝑋𝑡 + 𝑢𝑡 (10)

Following Leonte (2012) the necessary and sufficient conditions for sustainability will be

confirmed by testing whether 𝑋𝑡and 𝑅𝑡are cointegrated with 𝜑 = 1 and checking whether

the I(1) processes of 𝑋𝑡and 𝑅𝑡are cointegrated in equation (10). The null hypothesis to

be tested here is that the 𝜑 = 1 and 𝑢𝑡 is a stationary process. Hence, a necessary and

sufficient condition for sustainability is that 𝑋𝑡and 𝑅𝑡 are stationary and cointegrated variables of order one with the cointegrating vector being (1,-1) for the IBB to hold. Though, the proportionality between revenue and spending does not mean that R and X are equal, it only shows that they covariate by the same amount. Hence may suggest that budget balance in current periods implies no surpluses are generated to pay off old debts. The cointegration between revenue and spending provides a mechanism however, provide a mechanism to track fiscal stances of government over time. In addition to the cointegration test for sustainability, the Dynamic OLS (DOLS) method was also experimented with. The DOLS estimator technique is asymptotically equivalent to the Johansen’s (1988) maximum likelihood estimator and is considered appropriate in both large and small samples. The DOLS regression equation can be specified by making some

amendments to the equation 10, takes the following form:

𝑄𝑡 = 𝛿 + 𝜑𝑋𝑡 + ∑ 𝛾𝑡

𝑟

𝑖=−𝑟

∆𝑋𝑡 + 𝑢𝑡 (11)

Equation 11 is a standard augmented OLS regression models with addition of a few lead and lag differences of the regressor. According to Baharumshah and Lau (2007), using the DOLS estimation, a more efficient estimate of the coefficient of the cointegration

44

vector than by simple OLS can be derived. Under the DOLS approach, two alternative

hypotheses of fiscal sustainability can be tested:

(i) fiscal policy is sustainable if there is cointegration relationship between 𝑋𝑡 and

𝑅𝑡, with 0 < 𝜑 < 1; (ii) fiscal policy is unsustainable even if there is cointegration relationship between

𝑋𝑡 and 𝑅𝑡, with 𝜑 < 1 𝑜𝑟 𝜑 >1

Derivation of Fiscal Sustainability Index

The time series index for fiscal sustainability in line with Polito and Wickens (2005) is derived by rearranging equation 4 in a way to equate fiscal balance to the present value

of future primary deficit as represented by Equation 12.

𝐷𝑡 − lim𝑗→∞

1

(1 + 𝑟)𝑗+1𝐸[𝐷𝑡+𝑗+1] = ∑

1

(1 + 𝑟)𝑗+1

∞

𝑗=0

𝐸[𝑆𝑡+𝑗+1] (12)

. The fiscal sustainability index is then constructed by comparing the two sides of equation (12) based on the assumption that fiscal deficit and discount rate are endogenous and time-varying variables while the target level of the debt-GDP ratio is a choice. The index derived is then compared with the current debt-GDP ratio and n periods ahead with given fixed values of the deficit and discount rate. If, for example, the aim is to decrease discounted debt then the left-hand side should be negative, and the right-hand side gives the present value of the primary surplus required to achieve this reduction in debt. An increase in discounted debt requires a lower primary surplus. The measure of fiscal

sustainability is therefore based on an n-period horizon on the metric specified as follows:

𝐹𝑆𝑡,𝑛 = 𝐷𝑡 − lim𝑗→∞

1

(1 + 𝑟)𝑗+1𝐸[𝐷𝑡+𝑗+1] − ∑

1

(1 + 𝑟)𝑗+1

∞

𝑗=0

𝐸[𝑆𝑡+𝑗+1] ⋚ 0 (13)

Equation (13) above is the measure of fiscal sustainability proposed in this study. The index provides a comparison with the current level of the debt-GDP ratio. As 𝑗 → ∞ the

second term (i.e. lim𝑗→∞

1

(1+𝑟)𝑗+1 𝐸[𝐷𝑡+𝑗+1]) in 𝐹𝑆𝑡,𝑛 tends to zero and the index can be

interpreted as comparing the existing level of the debt-GDP ratio with the resources to pay it off hence we have:

𝐹𝑆𝑡,𝑛 = 𝐷𝑡 − ∑1

(1 + 𝑟)𝑗+1

∞

𝑗=0

𝐸[𝑆𝑡+𝑗+1] ⋚ 0 (14)

The NPG holds and fiscal policy stance is sustainable if at least 𝐹𝑆𝑡,𝑛 = 0 which implies

the debt-GDP ratio is forecast to be on target. But when only 𝐹𝑆𝑡,𝑛 > 0 then the

forecasted present value of the primary surplus is insufficient to achieve the desired change in the debt-GDP ratio. In this case the current fiscal stance is said to be

unsustainable.

3.3 Constructing an Economic Performance Index

One of the challenges in measuring performance is determining which macroeconomic variable best captures the overall economic situation and dynamics. While GDP has

45

remained popular, measure of fiscal performance, it failure to capture other measures of macroeconomic performance indicators led to developing a broader and more integrative

index for measuring macroeconomic performance.

To capture a more inclusive measure of macroeconomic performance, the paper followed the approach adopted by Khramov and Lee (2013). As stipulated by Khramov and Lee (2013), the economic performance index hereafter, (EPI) is a macroeconomic indicator that captures the overall performance of a country’s economy and reports any deviation from the desired level of economic performance. The EPI comprises variables that simultaneous influence the decision of the household, firms and government, this include: the inflation rate as a measure of the economy’s monetary stance; trade balance as a measure of the country’s external viability; the budget deficit as a percentage of total GDP as a measure of the economy’s fiscal stance; the change in real GDP as a measure

of the aggregate performance of the entire economy.

The calculation of the EPI score here is done annually by taking a total score of 100 percent and subtracting the inflation rate, the trade balance, budget deficit, and adding the percentage change in real GDP, all weighted and calculated as deviations from their desired values. Usually a grade is assigned to this score to further communicate economic

performance.

To begin with the construction of the index, an optimal EPI score is normalized to 100 percent and any score below 100 percent is defined as a decrease in economic performance. Next is a definition of the desired values for each of the sub-components

of the indicator which is as follows;

i. The desired rate of inflation is 0.0 percent ii. The desired trade balance is 0.0 percent iii. The desired level of government deficit as a percentage of GDP is 0.0 percent,

which is consistent with the long term balanced budget; iv. The desired change in GDP is a healthy real growth rate of 4.75 percent.

The deviations or convergence from/to these values show how an economy has performed holistically. The EPI score is constructed such that its value relates with the sub-indicators thus; falls when the inflation rate (IF) deviates from its desired value; rises when the trade balance (TB) is positive; falls when the government deficits (DF) rises from its desired values; and rises when the GDP (YD) growth rate is positive. The formula

for calculating the weighted EPI is given as follows:

𝐸𝑃𝐼 = 100 − 𝐼𝐹 − 𝐼𝐹∗) + (𝑇𝐵 − 𝑇𝐵∗) − (𝐷𝐹 − 𝐷𝐹∗) + (𝑌𝐷 − 𝑌𝐷∗) (15)

Where IF* is the desired inflation rate, TB* is the desired trade balance, DF* is desired fiscal deficit while the desired GDP I YD* . Equation 12 above can be rewritten in lower

case to denote the deviation thus;

𝐸𝑃𝐼 = 100 − 𝑖𝑓 + 𝑡𝑏 − 𝑑𝑓 + 𝑦𝑑, (16)

The EPI is calculated as 100 minus the absolute value of inflation, plus the trade balance, minus the budget deficit, plus the percentage change in real GDP all expressed deviations from their desired values. Khramov and Lee (2013) stipulates that changes in the economy affects the EPI in a straight forward manner. For example, if inflation rate

46

increases from 2 percent to 3 percent, the EPI falls by 1 percent and vice versa. Similarly, a 1 percentage point increase in deficit causes the EPI to fall by 1 percentage point. The opposite direction will be the case for a positive trade balance and GDP growth. Thus, while the former variables affect EPI negatively, the latter variables are positively related to the EPI. This paper deviates from the convention of examining debt and growth by regressing the stationary fiscal sustainability series on the new index derided from

equation 15 denoting macroeconomic performance1.

To provide a more concrete argument on the link between FFSI and EPI indices, the study carries out some empirical analysis including correlation, regression and causality. The regression result is done using an OLS method to obtain the empirical parameter estimates to show how EPI responds to FSI. This exercise is done by specifying a typical regression equation as follow;

𝐸𝑃𝐼𝑡 = 𝛼 + 𝛽1𝐹𝑆𝐼𝑡 + 𝛽2𝑌𝑡 + 𝑣𝑡 (17)

All the variables remain as earlier defined. 𝑌𝑡 is the growth rate in real GDP used to proxy the level of growth in economic activities in the economy. The inclusion latter two variable is also meant to reduce the estimation error that could arise from estimating a simple

OLS model and its attendant problems.

Data for the study are obtained from various sources including: World Development Indicators; central bank of Nigeria statistical bulletin for various years, and the national bureau of statistics. The data covers the period 1961 to 2016. The stationarity test, cointegration as well as other analysis are carried out on the logarithm of the fiscal sustainability variables discussed in the previous section. Using the logarithm of the variables wipes off any outline that may significant lead to spurious analysis

4. Empirical Results and Discussion

The process of establishing the fiscal sustainability in line with the framework above starts with, an integration (i.e. stationarity) test and a cointegration test between government revenues and expenditures is conducted. The estimation of the cointegration vector is used to establish sustainability of fiscal based on stipulated criterion. To examine the influence of fiscal sustainability on a new measure of macroeconomic performance, (i.e. calculating the coefficient of the fiscal sustainability variable), the paper employs the

dynamic ordinary least squares (DOLS) approach developed in Stock and Watson (2001).

Unit Root and cointegration Tests

The result of unit root tests is reported in table 1. All the variables were expressed in difference form after it was established that they were non stationary at levels. With this

result, it is established that the fiscal policy variables were established to be integrated of the same order I(I) as required. To establish if there exist any cointegration relationship between the fiscal policy variables. The Johansen’s (1995) cointegration test

and the unit-root test for the budget deficit𝑋𝑡 − 𝑅𝑡. The unit-root test for 𝑋𝑡 − 𝑅𝑡 is equivalent to the test of the cointegration relationship between𝐺𝑡, 𝑟𝑡𝐷𝑡 − 1, and 𝑅.

1 See Khramov and Lee (2013) for more detail discussion.

47

The first step in the cointegration test is to determine the optimal lag length. This is done with the aid of the SIC, AIC and FPE lag selection criteria. The result from the lag selection test shows that an optimal lag of 1 is appropriate to carry out the test. The result of the cointegration test is reported in table 3. The trace test statistic is used to evaluate the existence of cointegration based on the null hypothesis of no cointegration. The test is carried out under the linear deterministic trend specification which is to remain in tandem

with the theoretical model earlier outlined.

Table 1: Unit Root and cointegration Test Result

Augmented Dickey Fuller (ADF) Test Statistics

𝑿𝒕 𝑹𝒕 𝒓𝑫𝒕−𝟏

LAGS Ist. Diff. Ist. Diff. Ist. Diff.

0 -7.878** -10.15** -11.73**

1 -4.426** -6.172** -8.976**

2 -3.489* -4.050** -7.322**

Test of Cointegration Statistics

Hypothesized No. of CE Trace Statistic Critical Value Prob.**

𝑟 = 0∗ 66.85409 47.85613 0.0003

𝑟 ≤ 1∗ 33.38732 29.79707 0.0185

𝑟 ≤ 2 14.43633 15.49471 0.0717

𝑟 ≤ 3 1.621096 3.841466 0.2029

Note: The critical values of the ADF test at 0, 1 and 2 are 3.555, 3.557 and 3.546 respectively 𝑟 is the hypothesized number of cointegrating equations. * indicates rejection

of 𝐻0 of no cointegration This result from the cointegration test reported in the table 1 shows that there is at least one cointegration vector. The cointegration test supports a long run relationship between fiscal spending and revenue generation in Nigeria, thus indicating that there is potential for a long run fiscal sustainability in Nigeria. The Engel-Granger cointegration test was also carried out as a robustness test. The Engel-Granger cointegration test begins first by carrying out a regression such as specified in equation 10, thereafter the residuals from the regression are extracted and then the unit root is applied to this test. The application of the unit root test on the residual obtained from the estimation of equation 10 also showed that the errors in the estimation of revenue and expenditure converges back to an equilibrium value over a long-run horizon. The result of the Engel-Granger test is as reported in table 2.

Table 2: Engle Granger Cointegration Test Result

OLS Regression Result for 𝑹𝒕 and 𝑿𝒕

Engle-Granger Cointegration Test

Variable Coefficient t-Statistic Prob. ADF DF-GLS

Constant 0.019027 0.03631

6 0.9712 -9.3653 -3.7237

48

GEX 1.012367 56.6639

9 0.0000

D-W Statistics 0.901688 -6.6779 -1.9002

R-squared 0.98346

Adjusted R-squared 0.983154 -4.6579 1.0171

F-statistic 3210.808 0.0000

Test for Fiscal Sustainability Having established the result of the unit root and the Johansen and Engel and Granger cointegration test, the next course of action is to estimate equation (11) using the DOLS regression approach. The key coefficient in equation (11) is 𝜑 which measures the

response of the primary balance to the debt accumulation. The value of this coefficient is expected to be between zero and unity to be consistent with a stabilizing or sustainable fiscal policy response to rising debt. On the other hand, a negative or greater than unity coefficient denotes a destabilizing or unsustainable response. The result of the DOLS regression is reported in table 4. The regression is done using linear and quadratic trend specification. The DOLS is used here to estimate the cointegration vector and check whether the debt stock affects the sustainability of fiscal policy. The result from the DOLS regression confirms the case of fiscal sustainability and thus lends credence to that obtained earlier in the cointegration test. Based on the estimations, the fiscal policy responses to rising debt levels are entirely captured by simple linear decisions hence the inclusion of the non-linear trend. The estimated result suggests that fiscal balances in the country do respond in a stabilizing manner to increases in debt ratios. That is fiscal surpluses tend to increase in response to rising debt ratios. It can be inferred that fiscal surpluses tend to increase systematically to match rising debt accumulation. Hence, it can be said that the fiscal response to debt stock in Nigeria is sustainable. Though the coefficients of fiscal balance to current, lagged and squared debt ratios are small and statistically insignificant as denoted in the relatively marginal magnitude which include 0.019, -0.018 and 0.007 for current, lagged and squared debt ratios respectively, they are broadly in line with a few previous studies such Leonte, 2012; Franco and Zotteri, 2010; and Legrenzi and Milas, 2010 which also established lower response of fiscal balance to debt stock other developing countries’ economies in European and Latin American countries. Table 4: DOLS result for Fiscal Sustainability test

Variable Parameter Estimate T-stats. P-value

Dep. Variable Budget Balance 𝑄

Constant Term -0.72126 -0.250354 0.8036

Debt to GDP ratio (𝐷𝑡) 0.019073 0.921386 0.3622

Debt to GDP ratio (𝐷𝑡−1) -0.01878 -0.060842 0.9518

Squared Debt to GDP ratio (𝐷𝑡2) 0.007397 1.391293 0.1716

R-squared 0.7416

Adjusted R-squared 0.703786

The inclusion of the lag and squared debt trend provides information on how fiscal balance response to debt stock at different levels. This allows to check for the possibility of a non-linear response of fiscal balance to debt stock. On the overall, the results

49

suggested some evidence of “slight u-shaped” effects whereby the fiscal balance responds negatively to lag debt stock and later turns positively to positive debt stock. More importantly from the estimates, it is possible to set a limit to the extent of increase in debt profile, based on simple calculation, a limit of 28 per cent is established. This consistent with similar estimated derived earlier for Nigeria. The Debt Management Office in Nigeria sets a benchmark of 19.39 percent for the external debt ratio while the IMF set a benchmark of 56% for developing countries. Omotosho et al(2016) estimated a threshold of 73.7% for total debt/GDP ratio while a 30.9 per cent and 49.4 per cent for domestic and external debt respectively. The DMO estimates seems unrealistic in view of the fact since 2015, the debt ratio has consistently been above this less than 20 percent benchmark. The Omotosho et al (2016) estimate on the other hand portents danger for fiscal sustainability. Therefore, a conservative estimate of less than 30 per cent seems more realistic and encourage fiscal discipline Fiscal Sustainability and Economic Performance Having established the notion of fiscal sustainability in this study, the next objective is to examine the link or interaction between fiscal sustainability and economic performance. The next course of action in examining the link between fiscal sustainability and economic performance index. Figure 1 depict a graphical representation of the two series. This is meant to present a pictorial evidence and to see if a clear pattern of co-movement can be established between the two indices. The graph in figure 1 and 2 show some mild resemblance between the FSI and the EPI. The EPI index shows a somewhat topsy-turvy pattern of economic performance in the Nigerian economy. The graph shows that the first decade preceding independence the country recorded some progress in economic performance the country also maintained a positive value of FSI. This pattern remained

the same for most of the years in the decade preceding the end of the civil war.

Figure : Fiscal Sustainability and Economic Performance Indices in Nigeria (1961-2016)

-60

-50

-40

-30

-20

-10

0

10

20

30

40

50

19

61

19

65

19

69

19

73

19

77

19

81

19

85

19

89

19

93

19

97

20

01

20

05

20

09

20

13

FSI

-15.00

-10.00

-5.00

0.00

5.00

10.00

15.00

20.00

FSI EPI

50

Except for the period immediately preceding the end of the civil for most of the decade. However, the fiscal balance became unsustainable towards the tail end of the decade and continue into the early 1980s and this was attributed to the oil-glut that the resultant nose-diving of the price of the crude in the international market. Post-civil war era which was characterized by massive rehabilitation program of the government. On the other hand, the EPI shows a contradictory movement in the (1971-1980) depicting an unimpressive economic performance. The inconsistency in the stability of the FSI continued in the 1991 to 2000 episode. However, the case has been different, since the turn of the century and the return to democracy. This has been further attributed to the debt relief package the country obtained country and the adoption of major economic reforms by the government. It can also be alluded though intuitively that the positive FSI for this period may have been as result of the steady increase in the price of crude oil in the global market. As evidence in the graph, the FSI index persistently maintained a positive value throughout the period 2011 to 2016, though it dropped slightly. During the same period the EPI index also maintained the steady positive and significant value for

most of the years.

Thought the preceding evidence does not show a clear pattern of relationship between FSI and EPI a major point that can be deducible in this analysis is that except for the period immediately preceding the end of the civil war, the country’s unsustainable fiscal policy episodes are often occurring during periods of significant reduction in the international oil prices. This evidence well supports the analysis provided by Reinhart and

Rogoff (2011).

The result shows a weak but positive relationship with overall macroeconomic performance and level of economic activities in Nigeria. It shows specifically that EPI vary by approximately 0.42 for any unit change in FSI. The weak linkage notwithstanding, the estimate shows that macroeconomic performance improves as the government becomes more solvent and attain long run fiscal sustainability. As expected the parameter estimate for growth rate of real GDP takes on a positive and statistically value, thus indicating that the real GDP robustly affects EPI .The results also show that improvement in economic activities is key to sustaining a strong fiscal sustainability and sound macroeconomic performance in Nigeria. Although there is no theoretically stipulated or empirically established direction of the link between FSI and any EPI, the result is broadly in line with evidence obtained from previous studies (see IMF, 2003 and Adams, Ferrarini, and Park, 2010. The relevant statistics used to evaluate the soundness and consistency of the model are reported in the second part of table 5. The R-squared and Adjusted R-squared are low which is quite understandable in view of the nature of the model specified. In general, the soundness of the model is indicted by some of the statistic, however, this does not significant derail the result and conclusion derived here since the specification is not

guided by any theoretical framework

Table 4: Effects of Fiscal sustainability on Macroeconomic Performance Result

Variable Parameter Estimate T-Statistic Prob.

Dependent Variable: (EPI) Constant -2.29959 -0.72027 0.4745

LOG(FSI) 0.422848 0.953603 0.3446

LOG(GDPR) 0.9147** 3.571052 0.0008

51

R-squared 0.202904

Adjusted R-squared 0.172825

F-statistic 6.745682

D-W Stat 1.116747

Serial Correlation Test 6.8023** 0.0024

Homoscedasticity Test 0.127286 0.8808

Ramsey RESET Test 0.642981 0.4263

Note *** denote asymptotic significance at 1 per cent level.

5. Conclusion and policy implication

This paper attempts to provide empirical illumination to the argument of fiscal sustainability vis-à-vis economic performance in Nigeria. The study starts by providing the theoretical formulation of a simple tool: the long-run version of the government’s lifetime budget constraint. This tool links the size of the primary balance the government must run to maintain fiscal sustainability to the size of its debt, the flow of seigniorage it raises, the real interest rate. Using this formulation, the study articulates the conditionality for ascertaining if the Nigerian fiscal policy stance and activities has been sustainable. This study sets itself apart from previous studies by constructing a more elaborate index of economic performance and then using it to examine how economic performance

responds to dynamics in a calibrated fiscal sustainability index.

The empirical strategy for testing for fiscal sustainability begins by examining the unit root (i.e. stationarity) properties of government expenditure, government revenue and real interest rate using the ADF and DF-GLS unit root test framework. Next, the test for cointegration among the series is also carried using the Trace test developed by Johansen (1988) and the Engle-Granger cointegration test. To firmly concretize the outcome of the cointegration test, the DOLS method was used to estimate the cointegrating parameters inherent in the long run fiscal sustainability equation. Finally, the study also examines the

linkage between fiscal sustainability using some statistical and regression analysis.

On the overall, it can be inferred that the result from this study does not deviate from that of extant empirical studies. Evidence based on the unit root and cointegration test exercise suggest that fiscal policy is still within permissible sustainable range for Nigeria. The result from the DOLS estimation also lends support to that obtained from the unit root and cointegration test. The statistical analysis reveals a trend that has been well documented in the literature for other countries. Specifically, it shows that except for the period immediately preceding the end of the civil war, unsustainable fiscal policy episodes have largely been outcomes of significant decline in crude oil prices. The regression result for the responds of EPI to changes in FSI shows positive but only marginal respond which implies that significant asymmetry between government revenue and expenditure are

often reverted through counter-cyclical fiscal policy stance.

The main policy implication of this finding is that the Nigerian government should ensure a more robust and systemic link between tax and expenditures policies and the evolution of public debt. It must ensure that the fiscal debt/GDP ratio is always with the established realistic benchmark. However, the current debt/GDP benchmark adopted by debt office seems unrealistic and a more conservative estimate of 28 percent is suggested instead of the 19 percent debt/GDP ratio by DMO (2016) and 49 percent by Omotosho et

52

al(2006). In passing while political consideration always encourages fiscal expansion through increased borrowing especially during periods of economic downturn and revenue short-fall the long-term sustainability of such fiscal stance will depend significantly on the items such funds are expended on. If borrowing is used to finance capital projects such as productivity and efficiency enhancing infrastructure that creates wealth then this is sustainable, however, if funds are used for recurrent expenditure, this

may jeopardize the sustainability of fiscal policy.

Despite the results and conclusions reached in this study, there are still potential opportunities for extending the analysis presented here. A focus on determining a short-term government constrain framework and fiscal sustainability indicators for signaling short and medium term fiscal imbalances and to correct them will be a worthwhile direction for future research.

References Abdullah, H., Mustafa, M. M., & Dahalan, J. (2012). An empirical study on fiscal

sustainability in Malaysia. International Journal of Academic Research in Business and Social Sciences, 2(1), 72.

Adams, C., Ferrarini, B., & Park, D. (2010). Fiscal sustainability in developing Asia Afonso, A., Rault, C., (2009) 3-Step analysis of public finances sustainability: The case of

the European Union. Papers presented at the 65th Congress of the International Institute of Public Finance, Cape Town, South Africa.

Baharumshah, A. Z., & Lau, E. (2007). Regime changes and the sustainability of fiscal imbalance in East Asian countries. Economic Modelling, 24(6), 878-894.

Baharumshah, A. Z., & Lau, E. (2007). Regime changes and the sustainability of fiscal imbalance in East Asian countries. Economic Modelling, 24(6), 878-894.

Bohn, H. (2007). Are stationarity and cointegration restrictions really necessary for the intertemporal budget constraint?. Journal of monetary Economics, 54(7), 1837-1847.

Bohn, H., (2008). The sustainability of fiscal policy in the United States, in: Neck, R., J-E. Strum. eds, Sustainability of public debt. The MIT Press, Cambridge, Massachusetts, pp.15–49.

Bravo, A. B. S., & Silvestre, A. L. (2002). Intertemporal sustainability of fiscal policies: some tests for European countries. European Journal of Political Economy, 18(3), 517-528.

Burnside, C. (2004) Some Tools for Fiscal Sustainability Analysis. Deyshappriya, N. R. (2012). Debt and fiscal sustainability in Sri Lanka. International

Journal of Scientific and Research Publications, 2(3), 1-8. Fat´as, A., (2010). The economics of achieving fiscal sustainability. Franco, D., and Zotteri, F., (2010), Ensuring fiscal sustainability: Which role for fiscal

rules? The experience of European countries. Papers Presented at the 66th Congress of the International Institute of Public Finance, Uppsala, Sweden.

Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of economic dynamics and control, 12(2-3), 231-254.

Johansen, S. (1995). Identifying restrictions of linear equations with applications to simultaneous equations and cointegration. Journal of econometrics, 69(1), 111-132.

53

Khramov, M. V., & Lee, M. J. R. (2013). The Economic Performance Index (EPI): an Intuitive Indicator for Assessing a Country's Economic Performance Dynamics in an Historical Perspective (No. 13-214). International Monetary Fund.

Legrenzi, G., Milas, C., (2010) Spend-and-tax adjustments and the sustainability of the government’s intertemporal budget constraint. CESifo Working Paper No.2926.

Leonte, A., & Stoica, T. (2012). Assessing the effect of public spending on output in Romania using a value-at-risk (VAR) framework. African Journal of Business Management, 6(21), 6318.

Llorca, M., & Redzepagic, S. (2008). Debt sustainability in the EU New Member States: empirical evidence from a panel of eight Central and East European countries. Post-Communist Economies, 20(2), 159-172.

Mahmood, T., & Rauf, S. A. (2012). Public Debt Sustainability: Evidence from Developing Country. Pakistan Economic and Social Review, 23-40.

Miyazaki, T. (2014). Fiscal reform and fiscal sustainability: Evidence from Australia and Sweden. International Review of Economics & Finance, 33, 141-151.

Omotosho, B. S., Bawa, S., & Doguwa, S. I. (2016) Determining the Optimal Public Debt Threshold for Nigeria, Central Bank of Nigeria Journal of Applied Statistics. vol. 7, No. 2.

Polito, V., & Wickens, M. (2011). Assessing the fiscal stance in the European Union and the United States, 1970–2011. Economic Policy, 26(68), 599-647.

Prohl, S., & Schneider, F. G. (2006). Sustainability of public debt and budget deficit: Panel Cointegration analysis for the European Union member countries (No. 0610). Working Paper, Department of Economics, Johannes Kepler University of Linz.

Reinhart, C. M., & Rogoff, K. S. (2010). Growth in a Time of Debt. American Economic Review, 100(2), 573-78.

Rose, S., (2010) Institutions and fiscal sustainability. National Tax Journal, 63 (4), pp.807-838.

Silvestrini, A. (2010). Testing fiscal sustainability in Poland: a Bayesian analysis of cointegration. Empirical Economics, 39(1), 241-274.

Stock, J. H., & Watson, M. W. (2001). Vector Autoregressions. Journal of Economic perspectives, 15(4), 101-115.

Tanner E (2013) Fiscal Sustainability: A 21st Century Guide for the Perplexed International Monetary Fund WP/13/89

Tanzi, V. (2015). Fiscal and monetary policies during the Great Recession: a critical evaluation. Comparative Economic Studies, 57(2), 243-275.

Treasury (2013) Affording Our Future- Statement on New Zealand’s Long Term Fiscal Position- July-2013 http://www.treasury.govt.nz/government/longterm/fiscalposition/2013/affordingourfuture/ Treasury Newzealand Westerlund, J., & Prohl, S. (2010). Panel cointegration tests of the sustainability hypothesis in rich OECD countries. Applied Economics, 42(11), 1355-1364.

54

Capital Structure and Performance of Deposit Money Banks in Nigeria By Hafiz Usman Ahmed, Dr. Samaila I. Ningi & Dr. B.S. Dalhat 1. Introduction Capital structure is among the core decision areas in the field of finance. It determines the outstanding amount of debt and equity of a firm. It is an essential decision that has an intimate relationship with the value of firm (Paramasivan and Subramanian, 2008). This is due to the fact that capital structure has direct connection with the firm’s ability to fulfil the desires of its different stakeholders (Olokoyo, 2012). Capital structure also influences firm’s ability to deal with the competitive environment (Martis, 2013). This suggests that an optimal capital structure decision is essential to the firm’s survival (Ganiyu, 2015). Optimal capital structure represents the best combination of debt and equity that produce low cost of capital and maximises the firm’s value. Consequently, poor capital structure decision, for example wrong mix of debt and equity may lead high cost of capital, increase financial risk, lower the firm’s financial performance and eventually hinder its survival (Anarfo, 2015). This implies that inefficient capital structure

decision may force a firm to extinction.

Financial performance determines the firm’s efficiency in resources utilisation, in addition to its ability to make profit (Aymen, 2013). Financial performance of DMBs is paramount because of the critical role the industry plays in the economy in terms of provision of financial intermediation, transmission mechanism of monetary policy and maintenance of economic stability (Abbadi and Abu-rub, 2012). Stressing the importance of sound financial performance in the banking industry, Scott and Timothy’s study (as cited in Ronoh and Ntoiti, 2015) pointed out that banks with sound financial performance and sufficient capital can withstand adverse shocks. This implies that a sound banking sector will remain firm and continue to provide the needed financial intermediation services. Thus, a healthy and sound banking sector enhances financial deepening, creates more employment opportunities and promotes financial stability (Hafiz, 2018).

However, financial stability report indicates a declining trend in the Nigerian banks’ financial performance metrics. For instance, return on equity (ROE) and return on assets (ROA) have dropped from 14.90 and 2.67 percent in 2007 to 1.18 percent and 0.16 percent in 2016 (IMF, 2017; CBN, 2016). This development is capable of corroding public confidence and in the extreme, could trigger runs on the banks. Hence, the need for improvement in the key decision areas such as capital structure becomes imperative, because of their close relationship with the bank’s performance and survival.

55

The seminal exertion of Modigliani and Miller in 1958 formed the foundation of capital structure study. They posited that various combinations of debt and equity are irrelevant to the firm’s value. Afterward, the relationship between capital structure and financial performance has been studied substantially in both developed and developing countries with varying outcomes. Similarly, in Nigeria, there are studies on the relationship between capital structure and financial performance for both non-financial sectors (for example, Igbinosa, 2015 ; Chechet and Olayila, 2014; Arowoshegbe and Idialu, 2013; Uwoloma and Udiale, 2012; Onaolapo and Kajola, 2010) and financial sector specifically the banking sector that include the study of Idode, Adeleke, Ogunlowore, and Ashogbon (2014) on influence of capital structure on profitability of listed Nigerian banks, the impact of post-consolidation capital structure on the financial performance of quoted Nigerian banks by Adesina, Nwidobie and Adesina (2015), Uchechukwu and Kinsley (2016) regarding the effect of capital structure on firm performance of selected quoted banks in Nigeria, Shaba, Yaaba and Abubakar (2016) and Sadiq, Kachollom, Dasuki, and Yusuf, (2017). However, none of the prior studies from the context of banking sector consider different categories of debts such as short term debt (STD) and long term debt (LTD) as indicators of capital structure. Regarding the financial performance indicators, net interest margin (NIM) serves as a core banking specific performance indicator which has been to a large extent ignored in the previous studies. According to CBN (2013) margin metrics such as net interest margin (NIM) and net non-interest margin (NNIM) have recently turned out to be crucial indicators to banks’ management, regulatory authorities and the general public. Moreover, NIM shows the cost and efficiency of bank’s financial intermediation (Saksonova, 2014). This scenery therefore creates room for further studies so that

empirical evidences could be established from the Nigerian context.

Hence, this study intends to bridge the gap in knowledge by empirically assessing the robustness of the extent of relationship between capital structure metrics (short term debt ratio, long term debt ratio, total debt ratio and equity ratio) and financial performance of Nigerian banks using banks’ core business financial performance indicator, the net interest margin(NIM).

Subsequent to the introduction, Section 2 presents the literature review, empirical studies and theoretical framework. Section 3 presents methodology. In section 4 empirical results were presented. Finally, section 5 occupies the conclusion and recommendation.

2. Literature Review

In the literature review the paper presents categorisation of DMBs in Nigeria, conceptual and empirical review, theoretical framework and hypotheses development.

2.1 Categorisation of DMBs in Nigeria Following the repealed of the Universal Banking model in 2010, the new regulation provides that DMBs should operate within one of the following three categories (CBN, 2010): -

56

a. Regional Bank A bank in Nigeria with regional commercial banking authorisation license is permitted to conduct its commercial banking operations in at least six (6) states and at most in twelve (12) bordering States of the federation, spreading in not more than two (2) Geo-Political regions of the federation, plus the Federal Capital Territory. The banks that fall under this category are 2, the Suntrust Bank Nigeria Limited and Providus Bank Plc.

b. National Bank A bank in Nigeria with national commercial banking license is allowed to undertake its functions in all States of the federation including Abuja. There are 9 banks that fall in this category, these banks include but not limited to Citibank Nigeria Limited, Unity Bank Plc and Heritage Banking Company Limited.

c. International Bank A bank having international banking license is given the right to carry out its banking business operations in all States of the Federation, in addition it is also allowed to establish and maintain offshore banking operations in countries of its choice, subject to the CBN’s approval and the host country regulatory requirements. 10 banks fall under this stratum, for example Access bank, Zenith bank, Diamond bank, UBA and GTbank all are in this category.

The regulation also provides that the regional banks should have minimum paid up capital of N10 billion, N25 billion for the national banks and for banks with international authorisation license they should have a minimum paid up capital of N50 billion (CBN, 2010).

2.2 Conceptual and Empirical Review Capital structure is basically the way and manner in which a company finances its assets to generate income which invariably maximize the shareholders’ wealth. Saeed, Gull, and Rasheed (2013) opined that capital structure was indeed linked with different varieties of funding vehicles utilized by a company to get assets essential for its procedures as well as development. In the same direction Uwalomwa and Uadiale (2012) considered it as a mixture of company's long-term debt, specific short-term debt, common equity and preferred equity. Capital structure essentially depicts how a company funds its overall functions and growth by using diverse sources of funds. The company that is entirely financed by all equity is regarded as unlevered whereas firm that is financed with all debts is considered highly levered firm. However, it is not practically possible to finance firm entirely with debts in reality. Modigliani and Miller, (as cited in Chechet and Olayiwola , 2014) further stated that a firm that is all equity financed, the whole of its after-tax cash flows (profit) is a benefit to the shareholders inform of dividends and retained earnings. In contrast, company with certain proportion of debts in its capital structure shall devote a portion of the profit after tax to debt servicing (Chechet and Olayiwola, 2014). Hence, appropriate capital structure is closely related to the value of the firm (Tifow and Sayilir, 2015). In their study, Kundakchyan and Zulfakarova (2014) stated that an optimal mix of components of capital structure ensures corporate soundness, maximize return on capital and minimize financial risks. Conversely, the capital structure in banking sector is unique as compared to other business firms. Operationally, banks are financial intermediaries

57

that pool together money from surplus units and lend them to deficit units in the society (Kipesha and Moshi, 2014). In their study, Mostafa et al. (2011) opined that for banks to extend credit lines, entail mobilization of more funds such as acceptance new deposits, borrowing from other banks or equity issue. In confirmation of above assertion, Allen and Carletti (2013) contended that banks differ from other firms from the viewpoint of deposits mobilization. However, Miller (1995) opines that MM theory that formed the basis of capital structure theories can be applied to banks, basing his argument with the case of IBM lease financing subsidiary whose short term liability security “Variable Rate Book Entry Demand Note”, is functionally equivalent to demand deposits. Financial performance refers to financial metrics or indicators employed in determining the general well-being of a given entity. Bhunia, Mukhuti and Roy (2011) defined financial performance as firm’s overall financial health over a given period of time. The study added that analysis of financial performance is aimed at assessing the feasibility, solidity and fertility of a business. This implies that financial performance represents the result of firm’s operation in monetary terms for a specific period. Financial managers use ratios from company financial statement to assess its financial performance (Watson and Head, 2007; Bhunia,et al. 2011).One of the key factors used in measuring financial performance is profitability which according to Ross, Westerfield and Jaffe (2002) is the extent to which firm is able to generate profit from its operations. Profitability is the crucial objective of all business ventures; this is because the long run existence of these ventures depends upon their profitable operations. Its measurement is most remarkable indicator of business success (Khan, Sajid, Waseem and Shehzad, 2016). Samhan and Al-Khatib (2015)conduct a study on determinants of financial performance of Jordan Islamic Bank , covering the period year 2000 to 2012, return on assets (ROA), return on equity (ROE), and return on unrestricted investment accounts (ROUIA) were used to measured financial performance. Similarly, CBN in 2013 has buttressed the importance of the net interest margin (NIM) as an indicator of bank performance. Capital structure is an important decision for the survival and financial performance of banks because it affects the firm’s value. Debt and equity are the main components used by previous studies (for example, Sadiq et al., 2015; Ronoh and Ntoiti 2015) to measure capital structure of firms. In order to understand the relationship between capital structure and financial performance better, individual component of capital structure are discussed separately to outline how each component affects financial performance. Reaching a satisfactory debt level is critical for any business, not only because of the need to achieve profitability and firm value, but also because it increases an organization’s ability to deal with its competitive environment (Yazdanfar and Öhman , 2015). Debt capital is the money owed to others by the firm which must be repaid back within an agreed period of time (Kajirwa, 2015). Some varieties of debt instruments include but not limited to bonds and long-term notes payable (Siro, 2013). Furthermore, the use of debt capital may improve profit of an entity through shielding of tax ((Modigliani and Miller, 1963). In the same vein, debt capital increases the pressure on managers thereby motivating them to perform more efficiently. As a result, debt

58

financing reduces moral hazard behaviour by reducing free cash flow (Yazdanfar and Öhman, 2015). However, debt capital comes with a cost because interest on money borrowed needs to be paid as at when due, this increases firm’s financial risk (Kajirwa, 2015). Enekwe, Agu and Nnagbogu (2014) found that the amount of debts in the firm’s capital structure bears a negative insignificant relationship with the financial performance. This entails that firms do not assign much value to the debt financing for their growth. In a similar view, an empirical evidence provided by Sadiq, et al. (2017) have applied Pearson correlation coefficient and GLS regression model to examined the effect of capital structure on profitability of listed DMBs, the study found that capital structure has an effect on the financial performance of listed deposit money banks in Nigeria. The study recommends that deposit money banks in Nigeria should employ an appropriate mix of debt and equity capital. However, the study cannot be generalised due to its scope that limits its sample to four banks and possibility of spurious regression as shown by high R squared value of about 89%. Similarly, Shaba , Yaaba and Abubakar (2016) tudy the relationship between capital structure and profitability of deposit money banks in Nigeria. Applying autoregressive distributed lag model on a sample of 13 DMBs from 2005 through 2014, the study found that about 83 per cent of total assets employed by the DMBs are not financed by owners, confirming the hypothesis that banks are highly levered institutions. The results further found a positive and significant impact of both owners’ and borrowed funds on profitability proxied by gross earnings. Nevertheless, CBN (2013) has reinforced the importance of interest margins to account for the financial performance of banks in Nigeria which limits the study. Furthermore, Abubakar (2015) examined the relationship between financial leverage and financial performance, correlation technique used found an insignificant relationship between debt ratio and return on equity, and this indicates that the high debt ratio in the banks’ capital structure does not influence financial performance proxied by ROE. The major limitation of the study is the correlation method of analysis that was used to examine the causal effect instead regression, which is more appropriate technique. Awunyo-Victor and Badu (2012) in the study of Ghanaian banks found a negative relationship between leverage and return on equity at 10 percent level of significance. This implies that if banks decide to employ higher proportion of debt to finance their operations, their financial performance will reduce due to increase in the interest payment. Meaning that, an increase in the level of debt in the bank’s capital structure may result to high financial risk, and subsequently increases the risk of financial distress and bankruptcy. However, their result cannot be relied upon for the fact that conventionally the significance level for social and management sciences is 5 percent. Taani (2013) found that total debt ratio is a significant determinant of financial

performance of Jordanian banks disagrees with this proposition.

Bank access to equity capital perhaps has bearing on its ability to avoid bankruptcy cost (Aymen, 2013). Equity capital can be viewed from two dimensions (Aburime, 2008). These are the amount contributed by the owners of a bank (paid-up share capital) that gives them the right to enjoy all the future earnings and other funds available to support a bank’s business such as retained earnings and reserves. Equity capital is also termed as total shareholders’ funds. Bank’s equity capital is widely used as one of the determinants of bank profitability since it indicates the financial strength of the bank (Mungly et al., 2016). Furthermore, Aburime (2008) suggested that the bank level of

59

safety can be achieved by high capital level which could generate positive net benefits. Because banks with enough capital have the ability to absorb shocks from the problem of non-performing loans and provides a better shield to depositors in time of liquidation. Despite the role of equity capital as the most effective loss-absorption financial instrument, yet it possesses some social costs if it was achieved by holding back funds that are supposed to be granted as credit or through charging higher interest rates on credits (Oliver, Ruano and Fum´asc, 2013). Empirical evidence presented by Shaba, Yaaba and Abubakar (2016) evealed that the equity ratio which is the measure of the capital structure posted a positive relation with the banks financial Performance in Nigeria. This is however in contrast with the findings of Ronoh and Ntoiti, (2015). Prior studies such as Abubakar (2015) and Hailu (2015) have suggested the use of different metrics of capital structure and financial performance in future studies. The model of this paper therefore has been modified to capture different metrics of capital structure and financial performance. The metrics selected are short term debt ratio, long term debt ratio, debt ratio, and equity ratio to measure the independent variable. Whereas net interest margin (NIM) is to measure the dependent variable. These metrics are valid indicators of capital structure that were used by prior studies such as Abbadi and Abu-Rub (2012), Goyal, (2013), Taani (2013), Noor and Suardi, (2015) and Gebremichael (2016). Similarly, the financial performance indicator chosen was used in the following works (Naceur and Omran, 2011; Ongore and Kusa, 2013).

Following similar studies (for example, Yadav and Salim, 2012; Goyal, 2013; Anafo, Amponteng and Yin, 2015; Siddik, Kabiraj and Johgee, 2017), a set of control variables such as, bank size, and growth were selected in this paper. These control variables were deployed to avoid model misspecification and to hold constant some bank specific determinants of financial performance that may affect the result of the study.

2.3 Theoretical Framework Many theories relating to capital structure have been put forward by various scholars in the field of corporate finance. These include the work of Modigliani and Miller in 1958, which assumed that under the premise of perfect capital market various combinations of debt and equity are irrelevant to the firm’s value (Modigliani and Miller, 1958). Later, this assumption was relaxed to accommodate the effect of tax benefits on debt finance (Modigliani and Miller, 1963). Trade-off theory which assumes that firms trade off the benefits and costs of debt and equity financing and find an optimal capital structure after accounting for market imperfections such as taxes, bankruptcy costs and agency costs. Myers and Majluf (1984) in their pecking order theory argued that firms follow a financing hierarchy to minimize the problem of information asymmetry between the firm’s managers (insiders) and the (outsiders)shareholders or investors. Jensen and Meckling agency cost theory of 1976 suggested that, given an increasing conflict of interest between managers and the business owners, presence of more debt level in the firm’s capital structure imposes financial discipline, hence reduces agency problem. However, in order to connect capital structure and banks’ financial performance agency cost theory was adopted. The theory seems to be more relevant to the environment where laws are broken with impunity, capital market is inefficient and surrounded by several

60

imperfections, and corporate governance from the side of firm is weak. These features are aligning with the most emerging markets such as Nigeria. Olokoyo (2013) has also

used this theory in her study on the listed non-financial firms in Nigeria.

2.3.1 Agency Cost Theory Jensen and Meckling (1976) define the agency relationship inside the firm as: "A contract under which one or more person (the principal) engages another person (the agent) to perform some service on their behalf which involves delegating some decision-making authority to the agent”. According to this theory, the agent manager may pursue his personal objective or deliberately act in such a way that portrays lack of commitment, self-centeredness which may lead to firm losing its value significantly in contrast with the overall firm’s objectives that will maximizes its value. Consequently, conflict of interest may arise between the manager and the firms’ owners. Taking up more debt financing may reduce agency cost problems, apart from meeting up the expectation of shareholders, managers must strive hard to redeem the fixed obligation of debt. Therefore, managers are motivated to act in such a way that will protect their interest in terms of job security and welfare. Gansuwan and Önel(2012 ) added that debt engenders financial discipline. The agency cost theory backs a positive relationship

between capital structure and financial performance.

2.3.2 Trade off theory The trade-off theory was formed from the works of Kraus and Litzenberger (1973), Miller (1977), Scott (1977) and Kim (1978) among others. The theory suggests that firm’s capital structure depends on the tradeoff between the tax shield benefit of using debt and its attendant consequences in form of financial distress. The inconsequential gain from further debt declines with increase of debt proportion in firm’s capital structure, this also increases the marginal cost. Hence, for firm to achieve its overall value, tradeoff has to be central in choosing the proportion of debt and equity that it intends to use for financing its operation.

3. Methodology The research design adopted for this paper is panel design. The reason for this choice was that panel research strategy allows for repeated observations of some quantities about the same entities of study over time (Brooks, 2008). Similarly, according to Gujarati (2004) “panel design has advantage of more degrees of freedom, more efficiency and less collinearity”.

3.1 Sample and Data There are 21 DMBs in Nigeria as at December 2017, but the sample of banks to be included in this paper depends on the availability of data. For this reason, three filters were used to conveniently select the sample size. The filters are that the bank must be listed, not delisted and should have full length of data for the period. Furthermore, a listed company is expected to comply with the NSE’s requirement of financial disclosure. Hence, their financial reports are expected to be easily accessible and readily available. The result of this process displayed in Table 1, has produced 12 DMBs that accounts for 80 percent of listed banks population in Nigeria, 8 of these banks are from international stratum and 4 are from national stratum, as such their annual financial reports for 10

61

years covering 2007 to 2016 was used. In all the study has 120 observations or data

points making it a balanced panel study.

Table 1: Sample Size of the study

S/N Name Category NSE Status

1 Access Bank Plc International Listed 2 Diamond Bank Plc International Listed 3 Ecobank Nigeria Plc National Listed 4 Fidelity Bank Plc International Listed 5 First Bank of Nigeria Plc International Listed 6 First City Monument Bank Plc International Listed 7 Guaranty Trust Bank Plc International Listed 8 Stanbic IBTC Bank Plc. National Listed 9 Sterling Bank Plc National Listed 10 United Bank for Africa Plc International Listed 11 Unity Bank Plc National Listed 12 Zenith Bank Plc International Listed

Adapted and modified (CBN and NSE, 2017) 3.2 Method of Data Analysis Evidence from prior empirical studies showed that data was analysed using different approaches ranging from Spearman’s correlation, Ordinary Least Squares Regression (OLS), Panel Corrected Standard Error (PCSE) to establish the relationship between capital structure and financial performance (Abubakar, 2015; Amara and Aziz, 2014;

Sadikk et al, 2017).

Hausman test was conducted in order to choose the most appropriate panel estimation between fixed effect and random effect (Hausman, 1978). The test provides two estimates and compares the slope of their coefficients. The threshold is based on 5% level of significance, therefore if the P-value is greater than 5% , then the random effect model prevails otherwise fixed effect.The Hausman test result indicates X2 = 1.87; P>X2=0.9316, the P-value is greater than the 5% level of significance indicating that random effect model is the appropriate estimator than fixed effect. Similarly, the result of Breusch and Pagan Lagrangian multiplier test further validates the choice for random effects estimator with P-value of 0.000 which is less than 5 % level of significance, implying the presence of significant differences among the sampled listed

DMBs.

Similarly, post estimation regression diagnostic tests were conducted to ascertain the validity of the statistical inferences for the study and the results of normality and model specification tests are found to be favourable. However, the results of heteroskedasticity test and autocorrelation test are unfavourable. To remedy these problems an option of robust standard error was adopted in the estimation (Hoechle, 2007;Wooldridge, 2002,). Efficiency of estimator is generally improved by robust

standard error (Green, 2008).

62

3.3 Model Specification To estimate the relationship between each of capital structure indicator (STD, LTD, DR, ER) , with the listed banks’ financial performance (NIM) control variables i.e. (GR and BZ) were included in the model to isolate the effect of banks specific factors on financial performance indicator during the period. Hence, the mathematical expression of random

effect estimation model for this paper is presented below:

Yit = βX it + α +Uit + εit ....................................................................................(1)

Where Y denotes the dependent variable, β is the coefficient of independent variable X, α symbolises an intercept, U represents between entity errors, ε represents within entity error, i represents the cross-sectional units and t is the time period.

Thus the equation (1) becomes

NIM = β1STDit + β2LTDit+ β3DRit +β4ERit + β5LIQit + β6BSZit+ β7GRit + α +Uit + εit

Where: β1-β7 denote the coefficients, i represents the Nigerian banks (1-13), t is the time period of the paper (2007-2016). NIM: Interest earned on assets minus interest paid on borrowed funds divided by the

interest earning asset. STD: Short term debt to total assets. LTD: Long term debt to total assets. DR: Total debt to total assets. ER: Total Equity to total assets.

BSZ: Natural logarithm of total assets. GR: Assets of current year minus assets of previous year by the assets of current year In view of the above, this paper has proposed the following hypotheses: H1: There is no relationship between STD and NIM

H2: There is no relationship between LTD and NIM

H3: There is no relationship between DR and NIM

H4: There is no relationship between ER and NIM

4. Results and discussion The objective of this paper is to examine the relationship between capital structure and financial performance of listed banks in Nigeria. Thus, results of the analysis with the help of STATA (14) statistical software package are presented as follows: 4.1 Descriptive statistics of the data Descriptive statistics enable transformation of raw data into more meaningful information (Sekaran, 2003). To describe data in this paper, descriptive statistics such as, mean,

standard deviation, minimum and maximum were presented in Table 2.

63

Table 2: Descriptive Table (N=120)

Source: STATA (14) output, 2018

The descriptive statistics table shows the mean score of NIM is 0.054 for the sampled listed DMBs during the study period. This implies that, for every one naira invested in interest earning assets, listed DMBs earned about five (5) kobo out of it. The minimum NIM value recorded during this period was of 0.014 whereas 0.112 was its corresponding maximum value. Similarly, the standard deviation from the mean of NIM was 0.016. This shows that although some listed DMBs earned below average, there were some that earned about 12 kobo for every one naira invested in interest earning assets. Looking at the independent variable the listed DMBs capital structure measured by STD, LTD, DR and ER, they have mean values of 0.675, 0.171, 0.845 and 0.153 respectively. The maximum values of 0.879, 0.815, 0.972, 0.298 and minimum values of 0.073, 0.033, 0.678 and 0.028 respectively were also found. Implying about 85 percent of the listed DMBs’ capital structure was made of debt. Similarly, STD, LTD, DR and ER, deviate from their means on both sides by 0.117, 0.107, 0.053 and 0.051 respectively. From the side of control variables, the mean score of GR for the study period was 0.090, a maximum value of 0.497 and the minimum value was -0.184. Likewise, the value of GR can deviate from its mean by 0.114. Finally, bank size (BSZ) has an average value of 5. 975, while 5.164 and 6.632 values are for minimum and maximum respectively. BSZ can also deviate by 0.339. 4.2 Correlation matrix To determine the association between the entire variables of the study, correlation matrix was obtained as presented in Table 2. Similarly, correlation can be used to determine the presence of multicollinearity among the independent variables. Table 3: Correlation matrix(N=120)

NIM STD LTD DR ER GR BSZ

NIM 1.0000

STD 0.0519 1.0000

LTD -0.0440 -0.8936*

1.0000

DR -0.0772 0.2527* 0.0489 1.0000

ER 0.1047 -0.2467*

-0.0294 -0.9346*

1.0000

GR -0.4409*

-0.3155*

0.2851* 0.0288 -0.0559 1.0000

Variable Obs Mean Std. Dev Min Max

NIM 120 0.054 0.016 0.014 0.112 STD 120 0.675 0.117 0.073 0.879 LTD 120 0.171 0.107 0.033 0.815 DR 120 0.845 0.053 0.678 0.972 ER 120 0.153 0.051 0.028 0.298 GR 120 0.090 0.114 -0.184 0.497 BSZ 120 5.975 0.339 5.164 6.632

64

BSZ 0.0647 0.1910* -0.1157 0.1066 -0.0721 -0.2168*

1.0000

Correlation at 5%*, significance level Source: STATA (14) output, 2018 To determine the association among the variables of the study, correlation coefficients are obtained as presented in the correlation matrix table. The coefficients values revealed different levels of associations among the variables. For instance, net interest margin (NIM) exhibits a weak positive but insignificant association of 0.0519 with short term debt ratio (STD), negative insignificant correlation of -0.0440 with long term debt ratio (LTD), a significant negative association of -0.0772 with debt ratio (DR). In contrast, the correlation between NIM and equity ratio (ER) is 0.1047, suggesting a significant positive association. Similarly, the correlation matrix reveals that NIM has weak positive insignificant association of 0.0647 with bank size (BZ). Finally, a negative significant association of -0.4409 at 5 percent was found between growth prospects (GR) and NIM. Similarly, the extent of correlation among the independent variables was measured by the coefficient values. When the correlation between two independent variables is very strong, it is known as multicollinearity. And the implication of multicollinearity is that the multiple regression analysis cannot be relied upon. Conventionally, a correlation of more than 0.8 or less than -0.8 between two independent variables is a sign of multicollinearity (Garson, 2012). Most of the coefficient values were less than 0.8 and more than -0.8, with the exception of highest negative significant coefficient value of -0.9346 between debt ratio (DR) and equity ratio (ER). Similarly, the negative significant coefficient value of -0.8936 between STD and LTD was less than the required threshold value of -0.8. Hence, the need to further ascertain the extent of multicollinearity as signalled by the high correlation values with more robust technique. For this reason, variance inflation factor (VIF) was used in the context of this paper to examine the multicollinearity between the independent variables. According to Pallant, (2011) only VIF values that are greater than 10 and 1/VIF (Tolerance value) below 0.1 should be a cause for concern. Table 4: Variance Inflation Factor (VIF) Result

VARIABLE VIF 1/VIF

STD2 9.35 0.1069 DR 8.92 0.1121 LTD 8.43 0.1186 ER 8.03 0.1345 GR 1.18 0.8502 BSZ 1.09 0.9171

MEAN VIF 6.17

Source: STATA (14) output, 2018

The initial diagnostic test as indicated in the appendix has shown the presence of multicollinearity between STD, LTD and DR. To remedy the multicollinearity problem among the independent variables, one or more of the highly correlated variables must be

65

transformed as suggested by Hairs, et al. (2010). Table 4 above, shows the VIF and 1/VIF values of the study variables, long term debt ratio (LTD), short term debt ratio (STD), debt ratio (DR), equity ratio (ER), bank growth (GR) and bank size (BSZ) are all within the recommended caveat after the transformation. This implies the absence of multicollinearity among these variables. Hence, the regression estimation can be relied upon.

4.3 Regression Result Table 4: Random Effect Estimation Results

NIM Coeff. Robust Std. Err. Z

P> |Z| Significance

STD -0.0063 0.0155 -0.41 0.684

LTD -0.0126 0.0525 -0.24 0.811

DR 0.1002 0.0364 2.76 0.006 *

ER 0.1336 0.0268 4.99 0.000 *

GR -0.0587 0.0102 -5.78 0.000 *

BSZ 0.0007 0.0042 0.18 0.854

CONS -0.0507 0.0298 -1.70 0.089

Observations 120

R2 0.1907

Wald X2 19682.04 0.000 *

Hausman (X2>5%) 1.87 0.9316

Number of DMBs 12

Source: STATA (14) Output, 2018

Note: * Significant at 5% level of Significance

As indicated in the methodology, the model was analysed using random effect techniques with an option of robust standard error. The result presents R- squared value of 0.1907 for model, this indicates that capital structure indicators combined together with the control variables explained 19.07 % of the variability of listed DMBs’ financial performance (NIM). The Wald test value of 19682.04 for the model is significant at 5% and this provides an indication that this model is statistically fit to explain the listed banks’ financial performance (NIM) in Nigeria. The result is consistent with the findings of prior studies in Nigeria (for example Sadiq, Kachollom, Dasuki and Yusuf, 2017; Shaba, Yaaba and Ibrahim, 2016 ) and inconsistent with the findings of Anarfo (2015). Concerning the influence of each of the capital structure indicator with the NIM, the result found both STD and LTD have negative non significant relationship of (Coeff= -0.0063; P<z=0.684) and (Coeff= -0.0126; P<z=0.811) respectively. Going by the findings, the null hypotheses (H1 and H2) are accepted. This implies that an increase in STD or LTD has no significant relationship with a decrease in NIM and vice versa. The findings are consistent with Anarfo (2015). The remaining metrics DR and ER were however found to have positive and significant relationship of (Coeff= 0.1002; P<z=0.006) and (Coeff= 0.1336; P<z=0.000) with NIM respectively. With these results null hypotheses (H3 and H4) are rejected. Meaning, DR and ER are statistically determinants of NIM in this period. This finding is in line with the result of prior studies in Nigeria (for example Sadiq et al., 2017 and Shaba et al, 2016). Looking at the bank specific control variables, GR has a

66

negative significant relationship (Coeff= -0.0587; P<z=0.000). Growth in assets should have exhibit a positive relationship with financial performance. However, the negative result could be as a result of accumulated nonperforming or idle assets that these DMBs have, hence any increase in the GR could possibly erode their financial performance. BSZ posts a non-significant positive relationship (Coeff= 0.0007; P<z=0. 0.854). The finding suggests economies of scale do not play an important role in enhancing the NIM of listed banks in Nigeria in this period.

5. Conclusion and recommendations The study assessed the relationship of capital structure on the Nigerian bank’s financial performance. In view of this, the study observed that about 85% of the total capital of banks in Nigeria during the period of this study was made up of debt. This is reaffirmation of the fact that banks are highly levered financial institutions. The study found that capital structure indicators are good predictors of listed banks’ financial performance in Nigeria as evidence by the significant wald test value of less than 5 percent. Based on the findings obtained, the following recommendations are hereby offered. i. The result implies that profitable DMBs do not rely solely on STD to finance their

assets nor LTD. Therefore, bank management should consider a tradeoff between STD and LTD in making decision about capital structure in order to optimize their financial performance.

ii. Similarly, bank management should give more incentives to STD suppliers especially the depositors; this will motivate them to allow their deposits to stay with DMBs for a longer period than the present practice. The adjustment in maturity structure of STDs will provide DMBs with additional assets financing vehicle that could possibly enhance their performance. In addition, DMBs should desist from employing LTD only since it has a negative implication to their performance.

iii. Total debt (DR) is a significant determinant of listed banks’ financial performance and thus due diligence needs to be undertaken whenever bank decides to borrow funds for investment. This will ensure that managerial discipline enforced by debt on managers’ performance may not be outweighed by financial distress envisaged from excessive leverage.

iv. Equity contribution to net interest margin of the sampled listed DMBs is higher than debt contribution. This may not be unconnected with the high fixed interest obligation rooted with debt financing. Instead of over relying on debt financing as observed in descriptive matrix, banks managements should place emphasis on equity to finance their planned growth due to absence of fixed interest obligation.

v. There is need for the government to formulate policies that will fast track the development of a more vibrant capital market where DMBs and other firms will have access to equity and bond at global competitive rates. This will go a long way in

67

discouraging Nigerian firms from going offshore to seek financing opportunities and at the same time woos foreign investors to Nigerian capital market.

Reference Abbadi, S. M., and Abu-Rub, N., 2012, The effect of capital structure on the performance of Palestinian financial institutions. British Journal of Economics, Finance and Management Sciences, 3(2), pp92-101. Abubakar, A., 2015, Relationship Between Financial Leverage and Financial Performance of Deposit Money Banks in Nigeria. International Journal of Economics, Commerce and Management, 3(10), pp759-778. Aburime, U. T., 2008, Relationship of Political Affiliation on Bank Profitability in Nigeria. African Journal of Accounting, Economics, Finance and Banking Research, 4(4), pp61-75. Adah, A., 2012, Relationship of Recapitalisation on the Financial Performance of Deposits Money Banks in Nigeria (PhD Accounting and Finance), ABU, Zaria. Adair, J. T. A., 2006, Corporate Finance Demystified. New York The McGraw-Hill. Adesina, J., Nwidobie, B., and Adesina, O., 2015, Capital Structure and Financial Performance in Nigeria. International Journal of Business and Social Research, 5(2), pp21-31. Al-Tally, H. (2014). An Investigation of the Effect of Financial Leverage on Firm Financial Performance in Saudi Arabia’s Public Listed Companies. (Doctor of Business Administration), Victoria Melbourne, Australia. Al-Tamimi, H. H., 2010, Factors Influencing Performance of the UAE Islamic and Conventional National Banks. Global Journal Of Business Research, 4(2), pp1-10. Allahham, M., 2015, Relationship of Capital Structure on Bank Financial Performance of Al Ahli Bank in Saudi Arabia. Global Journal of Management and Business research: corporate Finance., 15(9). Allen, F., and Carletti, E., 2013, Deposits and Bank Capital Structure. FIC Working Paper, 13(13), pp1-58. Anafo, S.A., Amponteng, E. and Yin, L. (2015)Relationship of Capital Structure on Profitability of Banks Listed on the Ghana Stock Exchange. Research Journal of Finance and Accounting, 6 (16), 26-34.

Anarfo, E. B. 2015, Capital Structure and bank Performance - Evidence from Sub-Sahara Africa. European Journal of Accounting, Auditing and Finance Research, 3(3), pp.1-20.

68

Anonymous. 2015, List of Banks which shares are traded on the floors of Nigerian Stock Exchange. The Stalwart Report. Retrieved from The Stalwart report website: Arowoshegbe, A. O., and J.O. Idialu. 2013, Capital structure and profitability of quoted companies in Nigeria International Journal of Business and Social Research, 3(3), pp.99-106. Aslı, D.K., Martin, Č., Erik, F., Thorsten, B. and Ross, L. 2016, Financial Development and Structure Dataset , World Bank. Awunyo-Vitor, D., and Badu, J. 2012, Capital Structure and Performance of Listed Banks in Ghana. Global Journal of Human Social Science, 12(5), pp.56-62. Aymen, B.M. 2013, Relationship of capital on financial performance of banks: the case of Tunisia. Banks and Bank Systems, 84). pp.47-54. Ben Naceur, S., and Goaied, M. 2008, The determinants of commercial bank interest margin and profitability: Evidence from Tunisia. Frontiers in Finance and Economics, 5(1), pp.106-130. Bhunia, A., Mukhuti, S., and Roy, S. 2011, Financial Performance Analysis-A Case Study. Current Research Journal of Social Sciences, 3(3), pp.269-275. Brooks, C. 2008, Introductory Econometrics for Finance. New York: Cambridge University Press. Central Bank of Nigeria [CBN] 2017, list of deposit money banks in Nigeria. www.cebank.gov.ng CBN 2016, Financial Stability Report. www.cebank.gov.ng CBN 2015, Statistical bulletin. www.cenbank,gov.ng CBN 2013, Bank Intermediation in Nigeria: Growth, Competition, and Performance of The Banking Industry, 1990 – 2010. Occasional Paper No. 48 CBN 2010, CBN scope, conditions and minimum standards for Commercial Banks Regulations No. 01. Chechet, I., Garba, S., and Odudu, A. 2013, Determinants of Capital Structure in the Nigerian Chemical and Paints Sector. International Journal of Humanities and Social Science, 3(15), pp.247-263. Chechet, I., and Olayiwola, A. 2014, Capital Structure and Profitability of Nigerian Quoted Firms: The Agency Cost Theory Perspective. American International Journal of Social Science, 3(1), pp.139-158.

69

Chima, O. 2016, Financial Stability Report Reveals Decline in Banks’ Asset Quality. Retrieved 19th May, 2016 from http://www.thisdaylive.com/index. php/2016/05/19/ financial-stability-report-reveals-decline-in-banks-asset-quality/ Chinaemerem, O., and Anthony, O. 2012, Relationship of Capital Structure on the Financial Performance of Nigerian Firms. Arabian Journal of Business and Management Review (OMAN Chapter), 1(12), pp.43-61. Creswell, J. W. 2012, Educational research : planning, conducting, and evaluating quantitative and qualitative research (4th ed.). Boston: Pearson Education, Inc. de Bandt, O., Camara, B., Pessarossi, P., and Rose, M. 2014, Débats économiques et financiers N 12. Dlabay, L. R., and Burrow, J. L. 2008, Business Finance. USA: Thomson South-Western. Edralin, L. 2015, Capital Structure and Profitability of Selected Universal Banks in the Philippines Paper presented at the 9th Global Business Conference Manila, Philippines. Enekwe, C. I., Agu, I. C. and Nnagbogu, K.E. 2014, The Effect of Financial Leverage on Financial Performance: Evidence of Quoted Pharmaceutical Companies in Nigeria. Journal of Economics and Finance, 5(3). pp.17-25 . First Securities Discount House[FSDH]. 2016, Nigerian Banking Industry Report. Garson, D. 2012, Testing Statistical Assumptions. Ashiboro :Statisitcal Associates Publishing

Ganiyu, Y. O. 2015, Dynamic Analysis of the Relationship of Capital Structure on Firm Performance in Nigeria. (PhD), De Montfort, United Kingdom. Gansuwan, P., and Önel, Y. C. 2012, The Influence of Capital Structure on Firm Performance A quantitative study of Swedish listed firms (Master ), Umeå Sweden. Gebremichael, F. G. 2016, The Relationship of Capital Structure on Profitability of Commercial Bank of Ethiopia. Journal of Poverty, Investment and Development, 28, pp.17-36. Goyal, A. M. 2013, Relationship of Capital Structure on Performance of Listed Public Sector Banks in India. International Journal of Business and Management Invention, 2(10), pp. 35-43. Green, W. H. 2008, Econometrics analysis(6th ed.). New Jersey: Perason Prentice Hall. Gujarati, D. N. 2004, Basic Econometric 4th Ed., New York, McGraw-Hill Inc.

70

Hafiz, U. A. 2018, Impact of Capital Structure on Financial Performance of listed Deposit Money Banks in Nigeria. (MSc. Banking and Finance Thesis), Abubakar Tafawa Balewa University, Bauchi Hailu, A. 2015, The relationship of capital structure on profitability of Commercial Banks in Ethiopia.(M.Sc. Accounting and Finance Thesis), Addis Ababa, Ethopia. Hairs, J.F., Black, W.C., Babin, B.J. and Anderson, R.E. 2010, Multivariate Data Analysis (7th Edn.) Pearson Prentice Hall

Harding, J. P., Liang, X. and Ross, S. L. 2007, The Optimal Capital Structure of Banks: Balancing Deposit Insurance, Capital Requirements and Tax- Advantaged Debt. Economics Working Papers ; University of Connecticut. Harris, M and Raviv, A. 1991, Theory of Capital Structure. The Journal of Finance, 46(1),

pp. 297-355.

Hausman, J.A. 1978, Specification test in econometrics. Econometrica, 46: pp.1251-1271

Hoechle, D. 2007, Robust standard errors for panel regressions with cross-sectional dependence. The Stata Journal. 7 (3) pp.281-312.

Idode, P.E., Adeleke, T. M., Ogunlowore, A. J. and Ashogbon, O. S. 2014, Influence of capital structure on profitability: Empirical Evidence from listed Nigerian banks. OSR Journal of Business and Management. 16(11), pp.22-28. Igbinosa, S. 2015, Another Look at Capital Structure and Corporate Performance in Emerging Markets: The Case of Nigeria. Asian Journal of Business Management, 7(1), pp.1-12. Ikpefan O. A. 2012, Issues in Banking and Finance, CIBN Press, , pp.3 Imeokparia, L. 2015, Capital Base and Performance of Money Deposit Banks in Nigeria: Pre and Post Consolidation Era. International Journal of Managerial Studies and Research. 3(1), pp.74-82

IMF 2017, Nigerian Financial Sector Developments And Vulnerabilities. Country Report,No.12/81.

Jensen, M. and Meckling, W. H. 1976, Theory of the Firm: Managerial Behaviour, Agency Cost and Ownership Structure, Journal of Financial Economics, 3, pp. 305 – 360 Kajirwa, H. I. 2015, Effects of debt on firm performance: A Survey of Commercial Banks Listed on Nairobi Securities Exchange. Global Journal for Advanced Research. 2(6). pp.1025-1029. Khan, M., Sajid, M., Waseem, M., and Shehzad, M. 2016, Capital Structure Composition Demeanour towards Corporate Financial Performance Potential. International Journal of Innovation and Applied Studies, 14(1), pp.210-217.

71

Kim, E. 1978, A mean-variance theory of optimal capital structure and corporate debt capacity. Journal of Finance, 33,pp 45-63. Kipesha, E., and Moshi, J. 2014, Capital Structure and Firm Performance: Evidences from Commercial Banks in Tanzania. Research Journal of Finance and Accounting, 5(14), pp168-178. Kraus, A. and Litzenberger, R. H. 1973, A state-preference model of optimal financial leverage. Journal of Finance, 28, pp 911-922. Kundakchyan, R., and Zulfakarova, L. 2014, Current issues of optimal capital structure based on forecasting financial performance of the company. Life Science Journal, 11(6), pp. 368-371. Martis, R. N. 2013, Capital Structure and Firm’s Financial Performance: An Empirical Analysis of the SandP500. (Master Finance), Universitiet Van Tilburg, New Zealand. Miller, M. H. 1977, Debt and Taxes. Journal of Finance, 32(2), pp261-275. Mostafa, W., Eldomiaty, T., and Abdou, H. 2011, The effect of bank capital structure and financial indicators on CI’s financial strength ratings: the case of the Middle East. Banks and Bank Systems, 6(5), pp. 5-15. Mungly, Y., Seetanah, B., Seetah, K., Babajee, B. R., Maraye, P. N., and Ramdhany, N. G. 2016, Determinants of Mauritian Commercial Banking Profitability. . Paper presented at the Fifth Asia-Pacific Conference on Global Business, Economics, Finance and Social Sciences, Port Louis. Musa, H. M. 2015, Determinants o Financial Performance of Listed Mega Banks In Nigeria, M.Sc. Thesis Accounting and Finance, Zaria, Nigeria.

Myers, S. C. 1984, The Capital Structure Puzzle, Journal of Finance, 393): pp. 575-592. Myers, S. C., and Majluf, N. S. 1984, Corporate Financing and Investment Decisions when firms have information that investors do not have. Journal of financial economics, 13, pp. 187-221. Nguyen, H., and Kayani, Z. (2013, Determinants of Banks'capital structure in Asia: acomparison amongst developed and developing countries. (Master of Science in Business and Economics), Lunds Sweden. Nigeria Deposit Insurance Corporation (NDIC) 2017, http://www.ndic.org.ng/ Nigeria Stock Exchange (NSE) 2017, http://www.nse.org.ng/ Ningi, S. I., and Mairiga, T. H. 2015, The role of Bank of Inductry (BOI) in the Development of Small and Medium Scale Enterprises in Bauchi Metropolis. Research Journal of Finance and Accounting, 6(2), pp. 118-123.

72

Nirajini, A., and Priya, K. B. 2013, Relationship of Capital Structure on the Financial Performance of the listed trading companies in Sri Lanka. International Journal of Scientific and Research Publications, 3(5). Njeri, M. M., and Kagiri A. W. 2015, Effect of Capital Structure on Financial Performance of Banking Institutions Listed in Nairobi Securities Exchange .International Journal of Science and Research 4(7), pp. 924-930. Noor, K. D. and Suardi, I. 2015, The Relationship of Capital Structure on Financial Performance of the listed Agriculture Companies in Indonesia: Global Journal of Business and Social Science Review,3 (1), pp.310-318.

Nyor, T and Yunusa, A. 2016, Capital Structure and Operating Performance of Listed Conglomerate Firms in Nigeria. International Journal of Finance and Accounting. 5(2). pp.126-133. Okeke, C.T. 2005, Essentials of Business Finance. Bauchi: Multisys Nigeria Limited.

Oliver, A. M., Ruano, S. Fum´asc, V.S. 2013, Banks’ Equity Capital Frictions, Capital Ratios, and Interest Rates: Evidence from Spanish Banks. International Journal of Central Banking. 9(1). pp.183-225. Olokoyo, F. O. 2012, Capital Structure and Corporate Performance of Nigerian Quoted firms: A Panel Data Approach .(PhD. Thesis Finance), Ota, Nigeria. Oluduro, O. F. 2015, History and Evolution of Banking in Nigeria. Academia Arena, 7(1). pp.9-14 Onaolapo, A. A. and Kajola, S.O. 2010, Capital Structure and Firm Performance: Evidence from Nigeria. European Journal of Economics, Finance and Administrative Sciences. 25, pp. 70-82.

Ongore, V. O., and Kusa, G. B. 2013, Determinants of Financial Performance of Commercial Banks in Kenya. International Journal of Economics and Financial Issues, 3(1), pp.237-252. Onoja, I. R. 1998, The Contributions of Research Development in the Banking Industry in Nigeria. Nigerian Journal of Management Research. Pallant, J. 2011, SPSS Survival Manual: A step by step guide to data analysis using SPSS, (4th edn). Crows Nest. Allen and Unwin. Paramasivan, C., and Subramanian, T. 2008, Financial Management. New Delhi: New Age International Pvt Ltd Ronoh, C., and Ntoiti, J. 2015, Effect of Capital Structure on Financial Performance of Listed Commercial Banks in Kenya. A Case Study of Kenya Commercial Bank Limited. The Strategic Journal of Business and Change Management, 2(72), pp.750-781.

73

Ross S, Westerfield, R. and Jaffe, J. 2002, Corporate Finance, 6th Edition McGraw−Hill Primis Saeed, M. M., Gull, A. A., and Rasheed, M. Y. 2013, Relationship of Capital Structure on Banking Performance (A Case Study of Pakistan) Interdisciplinary Journal of Contemporary Research in Business, 4(10), pp.393-403. Sadiq. A.I., Kachollom, P.W., Dasuki, S. I. and Yusuf, M. 2017, Effect of Capital Structure on the Performance of Deposit Money Banks. International Journal of Accounting & Finance Review, 1(1) , pp.12-23. Saksonova, S. 2014, The role of Net Interest Margin in Improving Banks' Assets Structure and Assessing the Stability and Efficiency of their Operations. Procedia-Social and Behavioral Sciences. 150, pp132-141. Saliha, T., and Abdessatar, A. 2011, The Determinants of Financial Performance: An Empirical Test Using the Simultaneous Equations Method. Economics and Finance Review, 1(10), pp1-19. Samhan, H., and AL-Khatib, A. 2015, Determinants of Financial Performance of Jordan Islamic Bank. Research Journal of Finance and Accounting, 6(8), pp37-47. Scott, J. 1977, Bankruptcy, secured debt and optimal capital structure. Journal of Finance, 32, pp1-19. Scott and Timothy 2006 cited in Ronoh and Ntoiti 2015, Effect of Capital Structure on Financial Performance of Listed Commercial Banks in Kenya. A Case Study of Kenya Commercial Bank Limited. The Strategic Journal of Business and Change Management, 2(72), pp750-781. Shaba, Y., Yaaba, B. N and Abubakar, I. 2016, Capital Structure and Profitability of Deposit Money Banks: Empirical Evidence from Nigeria. European Journal of Business and Management, 8(23), pp110-121. Siddik, N.A.,Kabiraj, S. and Joghee, S. 2017, Relationships of Capital Structure on Performance of Banks in a Developing Economy: Evidence from Bangladesh.

International Journal ofFinancial Studies 5(13). pp.1-18.

Siro, R.O. 2013, Effect of Capital structure on financial performance of firms listed at the Nairobi Securities Exchange .(MBA. Accounting and Finance project), Nairobi, Kenya. Taani, K. 2013, The Relationship between Capital Structure and Firm Performance: Evidence from Jordan. Global Advanced Research Journal of Management and Business Studies, 2(11), pp.542-546. Tifow, A., and Sayilir, O. 2015, Capital Structure and Firm Performance: An Analysis of Manufacturing Firms in Turkey Eurasian Journal of Business and Management, 3(4), pp.13-22.

74

Uchechukwu, N. and Kingsley, A. M. 2016, Effect of Capital Structure on Firm Performance (A Study of Selected Quoted Banks in Nigerian Stock Exchange).The International Journal Of Business and Management. 4(4), pp. 114-122. Umoru,D. andOsemwegie, J. O. 2016, Capital Adequacy and Financial Performance ofBanks in Nigeria: Empirical Evidence Based on the Fgls Estimator. European Scientific Journal. 12(25). pp.295-305. Uwalomwa, U., and Uadiale, O. 2012, An Empirical Examination of the Relationship between Capital Structure and the Financial Performance of Firms in Nigeria. EuroEconomica, 31(1). Van Horne, J.C. and Wachowicz, J. M. 2008, Fundamentals of Financial management, 13th Ed. England. Pearson Education Limited.

Watson, D., and Head, A. 2007, Corporate Finance: Principle and Practice (4th ed.). Harlow, England. Prentice Hall Financial Times. Watson, D., and Head, A. 2010, Corporate Finance: Principle and Practice. (5th ed.). Harlow, England:: Prentice Hall Financial Times. Wooldridge, J. M. 2002, Econometric Analysis of Cross Section and Panel Data. Cambridge: The MIT Press.

Yadav, R. and Salim, M. 2012, Capital Structure and Firm Performance: Evidence from Malaysian Listed Companies. Procedia - Social and Behavioral Sciences, 65 (2012 ):

pp.156 – 166.

Yazdanfar, D. and Öhman, P. 2015, Debt financing and firm performance: an empirical study based on Swedish data. The Journal of Risk Finance. 16 (1). pp.102-118, Younus, S., Ishfaq, K., Usman, M., and Azeem, M. 2014, Capital Structure and Financial Performance: Evidence from Sugar Industry in Karachi Stock Exchange Pakistan. International Journal of Academic Research in Accounting, Finance and Management Sciences, 4(4), pp.272-279. Zeituna, R., and Tian, G. 2007, Capital Structure and Corporate Performance: Evidence from Jordan. Australasian Accounting Business and Finance Journal, 1(4), pp.24-36.

75

APPENDICES

76

NORMALITY TEST

010

20

30

40

De

nsity

-.05 0 .05Residuals

Kernel density estimate

Normal density

kernel = epanechnikov, bandwidth = 0.0038

Kernel density estimate

77

78

79

VARIABLES OF THE STUDY

SN CODE YEAR std ltd dr er bsz gr nim

1 1 2007 0.624 0.289 0.913 0.086 5.517 0.275 .036

2 1 2008 0.341 0.492 0.833 0.167 6.014 0.497 .025

3 1 2009 0.601 0.125 0.726 0.274 5.829 -0.184 .068

4 1 2010 0.606 0.143 0.749 0.251 5.862 0.032 .055

5 1 2011 0.553 0.251 0.804 0.196 5.976 0.114 .051

6 1 2012 0.722 0.121 0.843 0.157 6.181 0.205 .059

7 1 2013 0.714 0.142 0.856 0.144 6.231 0.051 .039

8 1 2014 0.668 0.193 0.862 0.138 6.297 0.066 .045

9 1 2015 0.634 0.217 0.851 0.149 6.382 0.085 .037

10 1 2016 0.586 0.278 0.864 0.136 6.491 0.108 .037

11 2 2007 0.678 0.150 0.827 0.173 5.495 0.146 .051

12 2 2008 0.669 0.137 0.806 0.194 5.781 0.286 .038

13 2 2009 0.683 0.138 0.821 0.179 5.814 0.033 .062

14 2 2010 0.691 0.096 0.787 0.213 5.739 -0.074 .09

15 2 2011 0.763 0.119 0.882 0.118 5.854 0.115 .097

16 2 2012 0.777 0.122 0.899 0.101 6.025 0.171 .08

17 2 2013 0.807 0.091 0.898 0.102 6.132 0.107 .073

18 2 2014 0.774 0.108 0.882 0.118 6.243 0.111 .058

19 2 2015 0.692 0.175 0.866 0.134 6.192 -0.051 .065

20 2 2016 0.683 0.190 0.873 0.127 6.221 0.029 .058

21 3 2007 0.073 0.815 0.888 0.112 5.493 0.372 .043

22 3 2008 0.718 0.208 0.927 0.073 5.636 0.143 .043

23 3 2009 0.686 0.108 0.793 0.207 5.551 -0.085 .065

24 3 2010 0.752 0.088 0.840 0.164 5.655 0.104 .063

25 3 2011 0.793 0.146 0.938 0.062 6.042 0.387 .026

26 3 2012 0.787 0.097 0.884 0.116 6.122 0.080 .046

27 3 2013 0.766 0.127 0.893 0.107 6.165 0.042 .061

80

28 3 2014 0.706 0.182 0.888 0.112 6.249 0.084 .056

29 3 2015 0.679 0.195 0.873 0.127 6.254 0.005 .069

30 3 2016 0.623 0.255 0.878 0.122 6.257 0.003 .069

31 4 2007 0.763 0.136 0.899 0.101 5.882 0.150 .05

32 4 2008 0.568 0.141 0.708 0.292 6.066 0.184 .049

33 4 2009 0.643 0.147 0.789 0.211 6.222 0.156 .052

34 4 2010 0.678 0.146 0.824 0.176 6.293 0.071 .058

35 4 2011 0.724 0.124 0.848 0.152 6.392 0.099 .072

36 4 2012 0.784 0.082 0.866 0.134 6.443 0.051 .074

37 4 2013 0.792 0.100 0.892 0.108 6.511 0.069 .064

38 4 2014 0.731 0.148 0.879 0.121 6.543 0.032 .062

39 4 2015 0.720 0.142 0.862 0.138 6.523 -0.020 .068

40 4 2016 0.700 0.163 0.863 0.137 6.551 0.028 .073

41 5 2007 0.715 0.167 0.882 0.118 5.420 0.392 .036

42 5 2008 0.541 0.175 0.716 0.284 5.668 0.248 .044

43 5 2009 0.627 0.125 0.752 0.248 5.711 0.044 .072

44 5 2010 0.632 0.114 0.746 0.254 5.724 0.013 .036

45 5 2011 0.692 0.110 0.802 0.198 5.773 0.049 .047

46 5 2012 0.711 0.143 0.855 0.145 5.958 0.185 .048

47 5 2013 0.709 0.148 0.857 0.143 6.004 0.045 .056

48 5 2014 0.628 0.235 0.863 0.137 6.068 0.064 .062

49 5 2015 0.604 0.256 0.860 0.140 6.064 -0.004 .055

50 5 2016 0.561 0.287 0.847 0.153 6.069 0.005 .059

51 6 2007 0.814 0.049 0.863 0.137 5.337 0.258 .041

52 6 2008 0.712 0.033 0.745 0.255 5.727 0.390 .041

53 6 2009 0.665 0.037 0.702 0.298 5.638 -0.089 .03

54 6 2010 0.685 0.034 0.719 0.281 5.679 0.042 .055

55 6 2011 0.759 0.057 0.816 0.184 5.869 0.189 .039

81

56 6 2012 0.784 0.040 0.823 0.177 5.961 0.092 .04

57 6 2013 0.746 0.103 0.849 0.151 6.034 0.073 .028

58 6 2014 0.691 0.163 0.854 0.146 6.074 0.041 .041

59 6 2015 0.625 0.226 0.851 0.149 6.091 0.016 .049

60 6 2016 0.611 0.246 0.857 0.143 6.113 0.023 .048

61 7 2007 0.608 0.293 0.901 0.099 5.680 0.195 .038

62 7 2008 0.485 0.319 0.804 0.196 5.963 0.283 .05

63 7 2009 0.649 0.166 0.815 0.185 6.009 0.046 .072

64 7 2010 0.668 0.140 0.808 0.192 6.028 0.020 .071

65 7 2011 0.632 0.213 0.845 0.155 6.183 0.155 .064

66 7 2012 0.651 0.173 0.823 0.177 6.210 0.026 .076

67 7 2013 0.663 0.164 0.827 0.173 6.280 0.070 .067

68 7 2014 0.677 0.154 0.831 0.169 6.328 0.048 .061

69 7 2015 0.625 0.197 0.822 0.178 6.357 0.030 .064

70 7 2016 0.646 0.172 0.818 0.182 6.417 0.060 .065

71 9 2007 0.238 0.524 0.762 0.238 5.484 0.439 .031

72 9 2008 0.286 0.492 0.778 0.222 5.539 0.055 .063

73 9 2009 0.514 0.259 0.773 0.227 5.521 -0.018 .073

74 9 2010 0.503 0.288 0.792 0.208 5.571 0.050 .069

75 9 2011 0.546 0.317 0.863 0.137 5.734 0.163 .053

76 9 2012 0.564 0.309 0.873 0.127 5.830 0.096 .05

77 9 2013 0.613 0.259 0.872 0.128 5.883 0.052 .049

78 9 2014 0.588 0.284 0.872 0.128 5.974 0.091 .05

79 9 2015 0.628 0.234 0.862 0.138 5.972 -0.002 .047

80 9 2016 0.584 0.283 0.866 0.134 6.023 0.051 .055

81 10 2007 0.733 0.084 0.816 0.184 5.164 0.124 .042

82 10 2008 0.781 0.091 0.872 0.128 5.374 0.210 .05

83 10 2009 0.780 0.112 0.892 0.108 5.313 -0.061 .061

82

84 10 2010 0.768 0.131 0.899 0.101 5.414 0.101 .056

85 10 2011 0.806 0.113 0.919 0.081 5.703 0.289 .029

86 10 2012 0.805 0.115 0.920 0.080 5.764 0.061 .041

87 10 2013 0.806 0.104 0.910 0.090 5.850 0.086 .051

88 10 2014 0.796 0.102 0.897 0.103 5.916 0.066 .052

89 10 2015 0.739 0.141 0.880 0.120 5.903 -0.013 .049

90 10 2016 0.686 0.211 0.897 0.103 5.919 0.017 .014

91 11 2007 0.814 0.036 0.850 0.150 6.042 0.112 .038

92 11 2008 0.828 0.049 0.876 0.124 6.182 0.140 .047

93 11 2009 0.822 0.044 0.866 0.134 6.146 -0.035 .077

94 11 2010 0.781 0.088 0.869 0.131 6.156 0.010 .044

95 11 2011 0.730 0.160 0.891 0.109 6.222 0.066 .034

96 11 2012 0.756 0.130 0.886 0.114 6.286 0.065 .039

97 11 2013 0.811 0.072 0.883 0.117 6.346 0.060 .034

98 11 2014 0.775 0.105 0.879 0.121 6.369 0.023 .035

99 11 2015 0.734 0.113 0.847 0.153 6.346 -0.023 .048

100 11 2016 0.669 0.177 0.846 0.154 6.405 0.059 .043

101 13 2007 0.717 0.125 0.842 0.158 5.308 0.191 .03

102 13 2008 0.879 0.069 0.948 0.052 5.561 0.253 .046

103 13 2009 0.837 0.135 0.972 0.028 5.410 -0.152 .065

104 13 2010 0.577 0.101 0.678 0.115 5.586 0.176 .043

105 13 2011 0.714 0.167 0.881 0.119 5.573 -0.013 .056

106 13 2012 0.682 0.188 0.870 0.130 5.597 0.025 .064

107 13 2013 0.751 0.179 0.930 0.070 5.606 0.009 .075

108 13 2014 0.670 0.145 0.815 0.185 5.616 0.010 .112

109 13 2015 0.522 0.292 0.814 0.186 5.647 0.030 .097

110 13 2016 0.536 0.295 0.831 0.169 5.693 0.046 .1

111 15 2007 0.643 0.230 0.872 0.128 5.946 0.162 .049

83

112 15 2008 0.691 0.107 0.799 0.201 6.225 0.279 .052

113 15 2009 0.706 0.085 0.791 0.209 6.197 -0.029 .066

114 15 2010 0.721 0.084 0.804 0.196 6.253 0.056 .047

115 15 2011 0.727 0.101 0.828 0.172 6.336 0.084 .056

116 15 2012 0.739 0.081 0.820 0.180 6.387 0.051 .061

117 15 2013 0.723 0.113 0.836 0.164 6.459 0.072 .065

118 15 2014 0.662 0.189 0.850 0.150 6.535 0.075 .054

119 15 2015 0.622 0.232 0.854 0.146 6.574 0.040 .054

120 15 2016 0.596 0.260 0.856 0.144 6.632 0.058 .049

Note:

Code Bank Name

1. Accesss Bank 2. Diamond Bank 3. Ecobank 4. FBN 5. FCMB 6. Fidelity Bank 7. Gtbank 8. Skye Bank *** 9. Stanbic IBTC 10. Sterling Bank 11. UBA 12. Union Bank *** 13. Unity Bank 14. Wema Bank *** 15. Zenith Bank

*** These banks were excluded from the analysis.

84

Effects of Switching Costs on Consumer Behavioural Intention in the Nigerian Deposit Money Banks

By: Ahmed Audu Maiyaki, Ph.D.

1.0 INTRODUCTION

In the service sector moving from one service provider to the other involves efforts, time and money; thus these serve as constraints to customer behavioural responses. Similarly, switching costs are regarded as powerful marketing tool used in influencing consumer behaviour (Klemperer 1995). Interestingly, Zeithaml (1981) found that the effects of switching barriers are more prevalent in consumer services. Such findings according to Burnham, Frels and Mahajan (2003) constitute a general consensus that switching costs are important factors in understanding consumer intention. Some scholars such as Farrell and Shapiro (1988) observed that switching costs enhance price inelasticity, and so they have been considered as tools employed by service providers to generate profit. Furthermore, Chang and Chen (2007) observed that switching barriers include hard or soft benefits provided by the firm. Hard benefits are economic gains that customers receive while soft benefits relate to the customer’s sense of ‘special status’ and recognition. They added that though various forms of switching barriers exist, some are more easily manipulated by the firm than others. Given these relationships it is vital for service providers to measure and monitor service quality and other antecedents of customer behavioural intention with a view to influence it. Globally, there seems to be expansion of service companies and organizations. This rapid growth of service industry is however, not limited to the developed economies; developing countries are equally experiencing a similar trend. According to Malthora, Ulgado, Agarwal, Shainesh and Wu (2005) and Vijayadurai (2008) developing economies in Asia, Latin America and Africa also have rapidly growing service sector that significantly contribute to the size of their respective GDPs. For example, Thailand’s economic survival is heavily dependent upon its tourism segment of the service sector (Nankervis, 2005). Given the above massive expansion of the global service sector, it is not surprising that it has attracted a lot of attention from both the researchers as well as industrialists. Some studies (e.g. Aydin, Ozer & Arasil, 2005; Caruana, 2004; Chang & Chen, 2007; Colgate & Lang, 2001; Goode & Harris, 2007; Wong & Mula, 2009; Xavier, 2008; Shukla, 2009) investigated the relationships between switching costs and behavioural intention; however, very limited published research specifically probe the role of switching cost as a moderating variable in the studies conducted so far. For instance, in his study Patterson (2004) examined the effect of switching costs as potential moderators of the link between satisfaction and retention, and they eventually found that switching costs/barriers are important factors influencing a customer's decision to remain with a focal service provider. Wong and Mula (2009) also investigated the moderating effect of switching cost in the relationship between satisfaction and retention. Considering the foregoing and the price sensitivity of Nigerian customers, again it is expected that switching costs could moderate the relationship between service quality and customer behavioural intention in the Nigerian banking industry. To this end, the main objective of the paper is to investigate the effects of service quality and switching costs on consumer behavioural intention. Especially since the issue of service charges has been among the major concerns of bank

85

customers in Nigeria (Anaro, 2009; Ochonma & Imoyo, 2008). The remaining sections of the paper consists of: Literature Review; Methodology; Results; Discussions of Findings; Conclusion; Managerial and policy Recommendations and References 2.0 LITERATURE REVIEW In this section related literature pertinent to the research have been critically reviewed and synthesized. This was done with a view to highlighting the studies and contributions made so far in the area under investigation. To this end, this research attempts to build and extent knowledge without duplicating research efforts already conducted. 2.1 Behavioural intention Several studies found that consumer behavioural intention can be either favourable or unfavourable depending on the nature of the factors determining the behaviour (Zeithalm, Berry & Parasuraman, 1996; Ladhari, 2009; Maiyaki & Mokhtar, 2011a). While favourable behavioural intention often leads to being loyal to the services provider, increased volume of transactions, expressing positive word-of-mouth, and a willingness to pay price premiums. On the contrary, unfavourable behavioural intention leads customers to display higher probability of brand switching, plan to reduce their volume of business, engage in negative word of mouth, and display an unwillingness to pay high prices (Zeithalm et al., 1996). Consequently, researchers observed that financial success and future performance of organizations depends on the extent to which customers favourable behavioural intentions are fostered (Dabholkar, Shepherd & Thorpe 2000). In the same vein, Malhotra and McCort (2001) argued that eliciting a greater understanding of consumers’ behavioural intentions continues to be a primary concern for marketing researchers. This is becomes obvious considering the frequency with which researchers have explored and modeled the antecedents of the behavioural intentions of consumers. Hence, customer perceived value, price and service quality have influence on customer bahavioral intention (Cronin, Brady & Hult, 2000). 2.2 Service Quality Studies have established that perception of service quality is influenced by both functional and technical dimensions (Chau & Ngai, 2010; Kang & James, 2004; Gronroos, 1982, 1990, 2001; Lehtinen & Lehtinen, 1982; Maiyaki, & Mokhtar, 2011; Parasuraman, Zeithmal, & Berry, 1985). According to Kang & James (2004) while functional dimensions of service quality focus on the service delivery process that is “how” the service is being provided, technical aspect focuses on the outcome quality of service delivery process that is “what” service is being provided. Hence, the process of quality evaluation occurs at the time when the service is being performed and is measured by the 5 dimensions of SERVQUAL instrument; on the other hand, technical quality evaluation takes place after service performance or rather the outcome of the service delivery. However, most of the research on service quality adopted SERVQUAL instrument which reflect only the functional dimensions of service quality and neglect the technical aspect (Gronroos, 1990; Kang & James, 2004). 2.3 Switching Costs It has been established that service quality is linked to customer switching behaviour and other behavioural outcomes such as complaint and recommendation to others (Yavas, Benkenstein, & Stuhldreier, 2004). Similarly, research suggested that different aspects of

86

service quality seem to be related to different outcomes; for example, tangible elements of service quality are more closely related to positive word of mouth; while ‘‘timeliness’’ aspects of service quality are more closely related to satisfaction and switching behavior (Yavas, Benkenstein, & Stuhldreier, 2004). Probably, the fairly large number of studies in consumer switching tendency is due to its relationship with the behavioural outcomes. Several researchers investigated switching behaviour in a variety of ways (Grewal, Iyer, & Levy, 2004; Jones, Mothersbaugh, & Beatty, 2000; Zeithaml, 1981). Consequently, Goode and Harris (2007) observed that quite different terms have been used to describe the switching tendencies such as switching costs and switching barriers. To this end, Colgate and Lang (2001) considered switching costs as a category of a broader switching barrier. However, for the purpose of this paper the terms are used interchangeably.

While there is a general agreement of what switching costs are, its constituents are however, less definite (Caruana, 2004). For example, Jones et al. (2000) described switching barriers as any factor that makes changing supplier (e.g. service provider) difficult or costly. Interestingly, Zeithaml (1981) found that these switching barriers are likely to be even more prevalent in consumer services. On the other hand, Goode and Harris (2007) argued that switching costs are often portrayed as a loss of financially quantifiable resources and consist of a financial loss to the customer. According to Klemperer (1995) consumers who have patronised an organisation tend to face some costs for switching to another firm even if the two organisations offered similar product/services. Hence, he argued that switching costs arises from a consumer’s desire for compatibility between his current purchase and a previous investment. Such costs may include; setting up the relationship, product/service usage and getting familiar with its features and psychological investment. In his study Patterson and Smith (2003) identified six switching barriers namely: search costs, loss of special treatment benefits, setup costs, risk perceptions, attractiveness of alternatives, and loss of social bonds. Similarly, Balabanis, Reynolds, and Simintiras (2006) itemised seven switching barriers in the online services. Notwithstanding the investments by the customer and the perceived switching barriers, some competitors offer incentives in order to attract potential customers. Similarly, there are a number of studies that investigated the effect of switching cost on behavioural intention; however, the concept has been tested with different set of variables in models that are although similar but different from the one focused in this study (e.g. Seo, Ranganathan & Babad, 2008; Aydin, Ozer & Arasil, 2005). Furthermore, Seo et al. (2008) drew the sample of his study from only one service provider and therefore, suggested that the result could be better reinforce with data from multiple service providers. Hence, this study investigates multiple retail banks in an attempt to fill in this methodological gap and thus, the following propositions are made: H1 Switching costs moderate the relationship between functional quality and the customer behavioural intention in the Nigerian commercial banks H2 Switching costs moderate the relationship between technical quality and the customer behavioural intention in the Nigerian commercial banks The foregoing hypotheses propose the moderating effect of switching cost on customer behavioural intention. More specifically, it is proposed that the higher the switching costs the more unfavourable customer behavioural intention towards the bank.

87

3.0 METHODOLOGY

Since the research has to use perceptions and opinions of the target respondents, a descriptive survey design was employed. Thus, primary data was gathered from the sample respondents using structured instrument. The total customer population of the selected clusters is one hundred and six thousand two hundred (106,200). Thus, using the table of sample size provided by Krejcie and Morgan (1970), a minimum sample size of 382 is required for the survey. Furthermore, in order to reduce sample size error and also to take care of the non-response problem, the required minimum sample size was doubled and rounded up to 800 (Hair, Wolfinbarger, & Ortinau, 2008). Furthermore, due to unavailability of sampling frame, convenient sampling technique was used to draw the respondents from the selected banks. To validate the research instrument construct validity and reliability were conducted. In order to test for the moderating effect, Multiple-group Structural Equation Modeling (MSEM) analysis using AMOS was employed. The moderating variable was re-coded into a categorical variable consisting of two groups i.e. high and low as suggested by Hair, Black, Babin, & Anderson, (2010) and Byrne (2010). This has been used a lot in marketing research (e.g. Jaworski & MacInnis, 1989). 3.1 Operationalisation and Measurement of Variables Switching Costs: While there are some subtle differences between switching “barriers” and “costs” (Goode & Harris, 2007), for the purpose of this study, the two terms are used interchangeably. The construct of switching costs/barriers can be described as the perception of the additional costs required for bank customers to terminate the current relationship with their banks and secure an alternative. The costs in question may either be monetary or non-monetary, such as psychological, emotional and the like. Switching costs was taken as a uni-dimensional construct for the purpose of this research as adapted from Colgate and Lang (2001); Caruana (2004); Aydin et al. (2005); Chang and Chen (2007); Goode and Harris (2007); Xavier (2008); Wong and Mula (2009). Behavioural Intention: This construct represents the dependent variable of the study and it was measured uni-dimensionally. Customer behavioural intentions depend on the service experience; the better the service experienced by the customer the more favourable his behavioural intentions would be towards the service provider and vice-versa. The variable serves as the first dependent variable to the antecedent factors while customer actual behaviour represents the second dependent variable. The measurement items were adapted from Zeithaml et al. (1996); Kang and James (2004); Tsaur et al. (2005); Olorunniwo et al. (2006); Vijayadurai (2008); Santonen (2007); Gu et al. (2009); Ladhari (2009). 3.2 Questionnaire Administration Furthermore, to get the sampling interval for the systematic selection, the number of customers was divided by the desired sample size for each bank, e.g., (150/54 ≈ 3). Hence, every third customer was selected to participate and in case he/she declines, then the next customer was chosen. Moreover, it was discovered that bank customers visit their banks more frequently on Mondays and Fridays. Similarly, the visits are higher at the beginning of each month. To this end, the researcher chooses to collect the data at the beginning and middle of the week and also the beginning and middle of the month. All the above procedures were followed in order to give every single element of the

88

population a known and equal chance of being selected. This would enable a representative sample that could be generalised to the target population. This is similar to the approach employed by Wan Omar (2008) and Hj Harun (2009). 3.3 Justification for Choice of Kano as Study area The fact that the population of customers of retail banks in Nigeria is fairly homogeneous, and that it is not part of the study’s objective to compare the banks’ of different regions, Kano metropolitan city was chosen as the area to be covered by the research. The city is one of the most commercially advanced cities in Nigeria. Kano has a location advantage as a centre of commerce and terminus of trade for centuries with some other African regions as well as the Arab world (Sani & Suleiman, n.d.). The study was restricted to Kano metropolis because of the concentration of commercial activities, which resulted in the presence of a large number of commercial banks in the city. All the commercial banks (national and international) operating in the country have at least one branch each in the city. Additionally, the metropolitan and commercial nature of the ancient city attracts people of different religions and ethnic backgrounds. 4.0 RESULTS

Although, composite reliability is stronger than the Cronbach’s alpha, in this study the latter was also assessed in order to complement the former. Cronbach’s alpha reliability can be judged by some rule of thumb criteria: alpha coefficient of more than 0.90 is categorised as excellent, the coefficient that is greater than 0.80 is classified as good, the one that is greater than 0.70 is acceptable, while the alpha coefficient which is greater than 0.60 is questionable, the alpha level of greater than 0.50 is categorised as poor for scientific research and finally, the alpha coefficient that is less than 0.50 is generally unacceptable (John & Reve, 1982). Generally, Cronbach’s alpha coefficient of 0.70 has been accepted as the minimum threshold for assessing reliability/internal consistency (Nunally & Bernstein, 1994). However, other authors like Hair et al. (2010) have recommended a lesser value of 0.60. From table 1 the values of Cronbach’s alpha for this study is between 0.854 and 0.943 and therefore, going by the above rule of thumb it could be concluded that the scales for this study have a high reliability standard (Hair et al., 2010; Nunally, 1967; Sekaran & Bougie, 2010). Factor loadings for all the items which ranged from 0.502 to 0.889, Confirming that the indicators are strongly related to their various constructs and hence, an indication of construct validity (Hair et al. 2010; Maiyaki, 2012). See appendix 1 for details. 4.1 Convergent Validity In an attempt to establish construct validity for this research, convergent validity was examined using Average Variance Extracted (AVE) as recommended by Hair et al (2010). This test shows how indicators of construct converged and how they share common variance. In other words, the indicators should converge and share high proportion of variance on a common point, the latent construct. AVE is computed as the mean of variance extracted for the items loading on a construct. This computation can be done using the formula bellow with the standardized loadings:

89

Where: Li = standardized factor loading i = number of items So from the formula AVE is the average of completely standardized factor loadings. Hence, according to Hair et al. (2010) an AVE of 0.5 is a good rule of thumb indicating adequate convergence; while an AVE of less than 0.5 suggests that, on average more error remains than variance explained by the latent factor structure imposed on the measure. It is obvious from the table, that all the constructs exhibit acceptable level of AVE at > 0.5 minimum criteria. The values of the AVE range between 0.515 and 0.765.

4.2 Discriminant Validity

Discriminant validity assesses the extent to which a construct is truly different from other constructs (Byrne, 2010; Hair et al., 2010). Consequently, high level of discriminant validity suggests that a latent construct is unique and captures some phenomena which other constructs do not. Although, there are several ways to compute discriminant validity, a more rigorous method is to compare the AVE values for any two constructs with the square of the correlation estimate between these two constructs. The AVE should be greater than the square correlation estimate (Hair et al. 2010). Another way of doing this test is compare the square-root of AVE for a given construct with the absolute correlations of that construct and all other constructs. In either way however, the AVE must be greater than the construct correlation in order to establish discriminant validity

(Fornell & Larcker, 1981). From the table 2 it is clear that all the square roots of AVE ranging between 0.739 and 0.874 are greater than the values of the constructs in the corresponding matrices. This therefore, indicates that each construct shared more variance with its items than it did with other constructs and thus, supporting discriminant validity.

4.3 Correlation Analysis

Correlation analysis was performed to establish association between the constructs under investigation. According to Mayer (1999b) relatively weak correlation of the range between 0.2 to 0.7 is sufficient enough to confirm the association between theoretically related constructs. From the above table 2, all the correlations between the variables are positive and significant at p < 0.01. It could also be seen from the same Table 2 that the correlations ranging between 0.425 and 0.799 are not weak and neither are they extremely strong (> 0.90) to indicate multicollinearity problem. The correlations are meaningful because construct that measuring similar phenomenon such as (functional quality vs technical quality) show relatively higher values. See appendix 2 for details.

Where:

i = loadings of indicator i of a latent variable

i = measurement error of indicator i j = flow index across all reflective measurement model

90

4.4 Hypotheses Testing From the AMOS output of multigroup SEM analysis under baseline model comparisons it was found that the ∆CFI between the unconstraint and constraints models for all the moderating variables (i.e. switching costs) is less than 0.01 and therefore passing the invariance test. The implication of this is that any subsequent changes in the multigroup models could be confidently attributed to the moderating effects of the variable under analysis. Hence, with the confirmation of invariance, the next logical step is to test the moderating effects. Table 3: Test of Moderation Effect

Moderating Effect of Switching Cost From Table 3 above it is clear that the X2 significantly changed from 1506.567 in the unconstrained model to 1550.971 in the constrained. The delta chi-square of 44.404 is greater the critical value of 31.41 and thus, considered significant at p < 0.05. Based on this result it could then be concluded that switching costs moderate the relationship between technical quality and behavioural intention. Furthermore, to evaluate the extent of significance between the groups of the moderating variable (high and low), the individual paths were analysed and the values compared (see Table 4 below). Thus, hypothesis 1 which proposes that switching costs moderate the relationship between functional quality and behavioural intention was not accepted. This is because from the statistical results, the path between functional quality and behavioural intention is not significant in both the structural model and multi group analysis. On the other hand, hypothesis 2 which states that “switching costs moderate the relationship between technical quality and customer behavioural intention” was accepted. Table 4 indicates that the moderating effect of high switching costs (β = 0.411**, t = 2.292) is stronger than the influence of low switching cost (β = 0.127, t = 1.221). Table 4: Comparisons of Path Coefficient and T-value (CR) for High and Low Switching Cost

Hypothesis Paths Hi cost Lo cost Comparison

Est. t-val Est. t-val

H1 Fqual Int 0.009 0.072 -0.225 -1.259 H = L

H2 Tqual

Int

0.411** 2.292 0.127 1.221 H > L

*** p < 0.001, ** p < 0.05, * p < 0.1 H = High Switching Cost

Model X2 Df RMSEA GFI CFI ∆X2 ∆d

f

Critical

Value

Sig

Basic model

921.164 419 0.047 0.9

0.960

Moderating variables

Costs Un-

constrained

1506.56

7

838 0.041 0.829 0.911

Constrained

1550.971

858 0.042 0.824 0.908 44.404

20 31.41 P < 0.05

91

L = Low Switching Cost Table 5: Summary of Hypotheses Test

Moderating Effects of Switching Costs H1 Functional Quality Behavioural Intention Not supported

H2 Technical Quality Behavioural Intention Supported

5.0 DISCUSSION OF FINDINGS Moderating Paths Moderating influence of switching costs on the relationship between the determinants of behavioural intention and customer behavioural intention

The objective of this paper is to examine whether or not switching costs could moderate the relationship between functional quality, technical quality, corporate image and perceived value on one hand, and customer behavioural intention on the other hand. To achieve this objective, four hypotheses have been proposed and tested as follows: In an attempt to find out the moderating effect of switching costs on the relationship between functional quality and behavioural intention, first hypothesis “switching costs moderate the relationship between functional quality and the customer behavioural intention in the Nigerian commercial banks” was tested using SEM multigroup analysis. The result shows that switching costs does not exert any significant moderating effect on the relationship under examination. From the output of multigroup analysis, both the high switching costs and the low switching costs have insignificant and weak beta and t-values with respect the path under analysis. This result is however not surprising given the fact that the path from function quality to behavioural intention was not significant in the revised structural model. Furthermore, in order to the assess whether switching costs moderate the relationship between technical quality and behavioural intention, hypothesis two “switching costs moderate the relationship between technical quality and the customer behavioural intention in the Nigerian commercial banks” was tested using the out of multigroup analysis from AMOS. As expected the results confirms that switching costs have significant moderating influence on the relationship in question. From the output, high switching cost has larger beta and t-values. This suggests that high switching costs exert greater moderating influence on the relationship between technical quality and behavioural intention. The implication of this finding is that since high switching costs has stronger moderating influence, meaning that when switching costs is high the relationship between technical quality and behavioural intention is stronger than when the switching costs is low. This corroborates Aydin et al. (2005) who found that the relationship between behavioural intention and its antecedent is stronger with high switching costs.

92

6.0 Theoretical Contributions

Given that virtually the studies reviewed which examine the moderating effect of switching costs, were conducted on other links different from the focus of this research. Consequently, the outcome of this research established that switching costs significantly moderate the causal relationship between technical quality and behavioural intention. The finding also confirms presumption that due to price sensitivity by the Nigerian retail bank customers, switching costs would have moderating effect on customer behavioural intention. To this end, this finding is hoped to significantly contribute to marketing theory.

7.0 Conclusion

Based on the foregoing research findings, the conclusions of this study are itemised as follows:

That the scale used in the research is valid and reliable. This is evident from the acceptable values obtained in terms of: Discriminant and Convergent validity, Composite and Cronbach’s Alfa reliability tests.

That switching costs have significant moderating influence on the relationship between technical quality and behavioural intention. Furthermore, high switching costs have stronger moderating effect on the relationship than low switching costs. On the contrary, switching costs do not have any significant moderating influence between functional quality and customer behavioural intention.

8.0 Managerial and Policy Recommendations

Moreover, from the empirical findings it has been discovered that switching costs positively affect the relationship between technical quality and behavioural intention. This finding further suggests that the effect of high switching costs is stronger on the relationships than the effect of low switching costs. Based on this therefore, the banks’ marketing executives and policy makers should strategise their marketing programmes in such a way that customer would relatively perceive high switching costs with regards to the bank’s services being delivered.

93

Reference

Anaro, B. (2009, November 18). Banking sector in The Gambia lies in Nigeria’s strength. BusinessDay. Retrieved from http://www.businessdayonline.com/index

Aydin, S., Ozer, G., & Arasil, O. (2005). Customer loyalty and the effect of switching costs as a moderator variable: A case in the Turkey mobile phone market. Marketing Intelligence and Planning, 23(1), 89-103

Balabanis, G., Reynolds, N., & Simintiras, A. (2006). Bases of e-store loyalty: Perceived switching barriers and satisfaction. Journal of Business Research, 59(2), 214–224.

Burnham, T.A., Frels, J.K., & Mahajan, V. (2003). Consumer switching costs: a typology, antecedents, and consequences. Journal of the Academy of Marketing Science,

31(2), 109-26.

Byrne, B. M. (2010). Structural Equation Modelling with AMOS: Basic concepts, application, and programming (2nd ed.). New York: Rouledge Taylor & Francis Group

Caruana, A. (2004). The impact of switching costs on customer loyalty: A study among corporate customers of mobile telephony. Journal of Targeting, Measurement and Analysis for Marketing, 12(3), 256-268.

Chang, Y., & Chen, F. (2007). Relational benefits, switching barriers and loyalty: A study

of airline customers in Taiwan. Journal of Air Transport Management, 13, 104–109

Chau, V.S., & Ngai, L.W.L.C. (2010). The youth market for internet banking services. Journal of Services Marketing, 24(1), 42–60

Colgate, M., & Lang, B. (2001). Switching barriers in consumer markets: An investigation

of financial service industry. Journal of consumer marketing, 18(4), 332-347

Cronin, J.J. Jr., Brady M.K., & Hult, T. M. (2000). Assessing the effects of quality, value and customer satisfaction on behavioural intentions in service environments. Journal of Retailing, 76(2), 193 – 218

Dabholkar, P.A., Shepherd, C.D., & Thorpe, D.I. (2000). A comprehensive framework for service quality: An investigation of critical conceptual and measurement issues

through a longitudinal study. Journal of Retailing, 76(2), 139-73.

Farrell, J., & Shapiro, C. (1988). Dynamic competition with switching costs. The RAND Journal of Economics, 19(1), 123-137.

Fornell, C., & Larcker, D. (1981). Structural equation models with unobservable variables and measurement error: Algebra and statistics. Journal of Marketing Research, 18(3), 382- 388.

Goode, M.M.H., & Harris, L.C. (2007). Online behavioural intentions: An empirical investigation of antecedents and moderators. European Journal of Marketing,

41(5/6), 512-536

94

Grewal, D., Iyer, G. R., & Levy, M. (2004). Internet retailing: Enablers, limiters, and

market consequences. Journal of Business Research, 57(7), 703–713.

Gronroos, C. (1982). Strategic management and marketing in service sector. Cambridge

MA: Marketing Science Institute

Gronroos, C. (1990). Service management and marketing. Lexington, MA: Lexington

Books

Gronroos, C. (2001). The perceived service quality concept- a mistake? Managing Service Quality, 11(3), 150-152

Hair, Jr., J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Uppersaddle River, New Jersey: Prentice Hall

Hair, Jr., J. F., Wolfinbarger, M. F., & Ortinau, D. J. (2008). Essentials of marketing research. Boston: McGraw-Hill/Irwin

Hj Harun, A. (2009). Culture of brand origin (COBO) and brand name linguistics their influences on attitude and purchase intention of a brand name: A cross cultural study. Unpublished doctoral dissertation, Universiti Sains Malaysia.

Im, I., Hong, S., & Kang, M. S. (2011). An international comparison of technology

adoption testing the UTAUT model. Information & Management, 48, 1–8

Jaworski, B. J., & Maclnnis, D. J. (1989). Marketing jobs and management controls:

Toward a framework. Journal of Marketing Research, 26(4), 406-419.

John, G., & Reve, T. (1982). The reliability and validity of key informant data from dyadic relationships in marketing channels. Journal of Marketing Research, 19(4), 517-

524

Jones, M. A., Mothersbaugh, D. L. & Beatty, S. E. (2000). Switching Barriers and

Repurchase Intentions in Services. Journal of Retailing, 76(2), 259–274

Kang, G.D., & James, J. (2004). Service quality dimensions: An examination of Gronroos’s

service quality model. Managing Service Quality, 14(4), 266-77.

Klemperer, P. (1995) Competition when consumers have switching costs: An overview with applications to industrial organization, macroeconomics and international trade.

Review of Economic Studies, 62, 515–39

Krejcie, R. V., & Morgan, D. W. (1970). Determining sample size for research activities.

Educational and Psychological Measurement, 30, 607-610.

Ladhari, R. (2009). Service quality, emotional satisfaction, and behavioural intentions: A

study in the hotel industry. Managing Service Quality. 19(3), 308-331

Lehtinen, U., & Lehtinen, J.R. (1982). Service quality: A study of quality dimensions. Unpublished working paper, Helsinki: Service Management Institute, Finland OY.

95

Maiyaki, A. A. & Mokhtar, S. S. M. (2011). Determinants of Customer Behavioural Intention in Nigerian Retail Banks. Interdisciplinary Journal of Research in Business. Vol. 1, Issue, 10, pp. 42- 48

Maiyaki, A. A. (2012). Influence of service quality, corporate image and perceived value on customer behavioural responses: CFA and Measurement Model. International Journal of Academic Research in Business and Social Sciences, Vol. 2,

No. 1, pp. 403- 414

Maiyaki, A. A. & Mokhtar, S. S. M. (2012). Determinants of Customer Behavioural Responses in the Nigerian Retail banks: Structural Equation Modeling Approach. African Journal of Business Management. pp. 6(4), 1652-1659. doi:

10.5897/AJBM11.2335

Malhotra, N.K., & McCort, J.D. (2001). A cross-cultural comparison of behavioural

intentions Models. International Marketing Review, 18(3), 235-69.

Malhotra, N.K., Ulgado, F.M., Agarwal, J., Shainesh, G., & Wu, L. (2005). Dimensions of service quality in developed and developing economies: Multi-country cross- cultural comparisons. International Marketing Review, 22(3), 256- 278

Mayer, R. E. (1999b). Research-based principles for the design of instructional messages.

Document Design, 1, 7–20.

Nankervis, A. (2005). Managing Services. New York: Cambridge University Press

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New-York:

McGraw-Hill

Ochonma, M., & Imoyo, I. (2008, November 3). FCMB sets new standard in customer relationship. BusinessDay. Retrieved from http://www.businessdayonline.com/index

Olorunniwo, F., Hsu, M.K., & Udo, G.J. (2006). Service quality, customer satisfaction,

and behavioural intentions. Journal of Services Marketing, 20(1), 59–72

Parasuraman, A., Zeithmal, V., & Berry, L.L. (1985). A conceptual model of service quality

and its implications for future research. Journal of Marketing, 49(4), 41-50.

Patterson, P. G., & Smith, T. (2003). A cross-cultural study of switching barriers and propensity to stay with service providers. Journal of Retailing, 79(2), 107–120

Patterson, P.G. (2004). A contingency model of behavioural intentions in a services context. European Journal of Marketing, 38(9/10), 1304-1315

Sani, B.M., & Sulaiman S. (n.d.). The Structure of Kano Economy. Retrieved from

http://www.kanoonline.com/

Santonen, T. (2007). Price sensitivity as an indicator of customer defection in retail

banking. International Journal of Bank Marketing, 25(1), 39-55

Sekaran, U., & Bougie, R. (2010). Research methods for business: A skill building approach (5th ed.). Chichester: John Willey & Sons Ltd

96

Seo, D., Ranganathan, C., & Babad, Y. (2008). Two-level of customer retention in the US mobile telecommunications service market. Telecommunications policy, 32, 182-

196

Shukla, P. (2009). Impact of contextual factors, brand loyalty and brand switching.

Journal of Consumer Marketing, 26(5), 348–357

Tsaur, S., Lin, C., & Wu, C. (2005). cultural differences of service quality and behavioural intention in tourist hotels. Journal of Hospitality & Leisure Marketing, 13(1), 41-63

Vijayadurai, J. (2008). Service quality, customer satisfaction and behavioural intention in

hotel industry. Journal of Marketing & Communication, 3(3), 14-26

Wan Omar, M. B. (2008). Understanding the mediating effect of customer satisfaction on the relationship of sales person behaviour performance and customer loyalty. Unpublished doctoral dissertation, Universiti Sains Malaysia.

Wong, C., & Mula, J.M. (2009). The moderating effect of switching costs on the customer satisfaction-retention link: retail internet banking service in Hong

Kong. IBIMA business review, 2, 20-28

Xavier, P., & Ypsilanti, D. (2008). Switching costs and consumer behaviour: Implications for telecommunications regulation, 10(4), 13-29.

Yavas, U., Benkenstein, M., & Stuhldreier, U. (2004). Relationships between service quality and behavioural outcomes: A study of private bank customers in Germany. The International Journal of Bank Marketing, 22 (2/3), 144-157

Zeithaml, V.A. (1981). How Consumer Evaluation Processes Differ between Goods and Services. In J. Donnelly and W. George, (Eds.), Marketing of Services (pp. 186- 190). Chicago: American Marketing

Zeithaml, V.A., Berry, L.L., & Parasuraman, A. (1996). The behavioural consequences of

service quality. Journal of Marketing, 60(2), 31-46.

97

Appendix 1: Constructs Validity and Reliability

Constructs Items Factor Loadings

Factor Loadings

Squared

Average Variance Extracted

Composite Reliability

Cronbach’s Alpha

Functional Quality

Tan1 0.831 0.691

0.609 0.959 0.943

Tan2 0.502 0.252

Tan3 0.777 0.604

Rel2 0.814 0.663

Rel3 0.731 0.534

Ass1 0.800 0.640

Ass3 0.778 0.605

Res1 0.746 0.557

Res2 0.780 0.608

Res3 0.822 0.676

Res4 0.822 0.676

Emp1 0.808 0.653

Emp2 0.760 0.578

Emp3 0.850 0.723

Emp5 0.824 0.679

Technical Quality

Tech1 0.839 0.704

0.690

0.899

0.943

Tech2 0.869 0.755

Tech3 0.837 0.701

Tech4 0.776 0.602

Pval6 0.826 0.682

Switching Cost

Cost2 0.716 0.513

0.596

0.855

Cost4 0.749 0.561

0.854 Cost5 0.805 0.648

Cost6 0.814 0.663

Behavioural Intention

Wom1 0.889 0.790

0.765

0.907

0.907 Wom2 0.868 0.753

Wom3 0.867 0.752

98

Appendix 2: Correlation Analysis

Tech Scost Int Fqual

SQUARE ROOT OF AVE

.831 .772 .874 .780

Tech Pearson Correlation

1

Sig. (2-tailed)

N 555

Scost Pearson Correlation

.425** 1

Sig. (2-tailed) .000

N 555 555

Int Pearson Correlation

.590** .561** 1

Sig. (2-tailed) .000 .000

N 555 555 555

Fqual Pearson Correlation

.799** .432** .614** 1

Sig. (2-tailed) .000 .000 .000

N 555 555 555 555