See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/341521330

IMPACT OF MONETARY POLICY AND ITS LAG ON ECONOMIC GROWTH (1986-

2016) (A COMPARATIVE STUDY OF U.K., U.S.A. AND NIGERIA)

Thesis · May 2020

CITATIONS READS

0 286

1 author:

Kehinde Miracle Alabi

Ekiti State University, Ado Ekiti
5 PUBLICATIONS   0 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Impact of Monetary Policy and its Lag on Economic Growth: A Comparative Study of U.K., U.SA. and Nigeria (1986-2016) View project

All content following this page was uploaded by Kehinde Miracle Alabi on 20 May 2020.

The user has requested enhancement of the downloaded file.

 IMPACT OF MONETARY POLICY AND ITS LAG ON
ECONOMIC GROWTH (1986-2016)

(A COMPARATIVE STUDY OF U.K., U.S.A. AND NIGERIA)

BY

ALABI, KEHINDE MIRACLE

1201798

BEING A THESIS SUBMITTED TO

THE DEPARTMENT OF BANKING AND FINANCE,
FACULTY OF MANAGEMENT SCIENCES,
EKITI STATE UNIVERSITY, ADO EKITI, EKITI STATE,
NIGERIA.

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR

THE AWARD OF BACHELOR OF SCIENCE DEGREE
(B.Sc. Hons.) IN BANKING AND FINANCE.

NOVEMBER, 2017
 DECLARATION

I ALABI, Kehinde Miracle hereby declare that the work contained in this thesis was originally

written by me and that the same has not been submitted by any other person for any award of

degree whatsoever.

……………………………….. ……………………………………
ALABI, Kehinde Miracle Date

2
 CERTIFICATION

I certify that this thesis was carried out by ALABI, KEHINDE MIRACLE with Matric.

Number 1201798 in the Department of Banking and finance as part of the requirements for the

award of Bachelor of Science (B.Sc.) in Banking and finance, Faculty of Management Sciences,

Ekiti State University under my supervision.

……………………………….. ……………………………………
Dr. S. O. Dada Date
(Supervisor)

……………………………….. ……………………………………
Dr. M. O. Oke Date
(Head of Department, BFN)

3
 DEDICATION
This thesis is dedicated to the Lord GOD Almighty, the solid pillar of my life, to the

blessed memory of my late grandfather, Chief John Oni Alabi, and to my parents, Pastor W. O.

and Pastor (Mrs.) J. O. Alabi.

4
 ACKNOWLEDGEMENTS

The hands that lifted me shall uphold me to the end and in view of this, my profound

gratitude goes to my father in heaven for his hands, unfailing love and grace upon my life; for He

has been the source of my sustenance and has furnished me with the needed strength in my

academic pursuit. I am be eternally indebted and forever grateful to Him.

My sincere gratitude also goes to my amazing supervisor Dr. S. O. Dada for his

unflinching support, coaching and mentorship throughout the programme. I am grateful for the

opportunity given to me to research and gain more knowledge on my project topic. I would also

like to appreciate all my lecturers in the department, faculty and the university at large including

those I came in close contact with during the course of my programme the list are numerous but

to mention a few are Prof. J. A. Oloyede, Prof. Popoola, Dr. S. O. Adeusi, Dr. A. O. Adaramola,

Dr. M. O. Oke, Dr. (High Chief) L. B. Ajayi, Dr. L. A. Sulaiman, Dr. (Mrs.) B. A. Azeez (a

mother indeed), Mr. F. T. Kolapo, Mr. J.O. Mokuolu, Dr. S. O. Dada, Mr. A. A. Obalade,

Mrs. G. O. Obisesan, Mrs. A. O. Adejayan, Prof. Adebayo (DSA), Dr. J. T. Owolabi, Dr. Ben

Bankole, Prof. Femi Adeoluwa, Mr. Adedapo Adesina and non-teaching staff members who

have at one point or the other impacted knowledge into me.

My profound gratitude goes to my unique and matchless family members who mean a lot

to me for their genuine love and support. These lovely people include Pastor W. O. Alabi (my

exceptional dad), Pastor (Mrs.) J. O. Alabi (my loving mum), Taiwo Alabi (LL.B. in View-My

Twin Brother) and Precious Alabi (my younger brother). My special thanks goes to my family

and adult friends such as Areola Iyanu (Nee Olugbade) who is a pacesetter for me in this field

and provided me with most materials I needed to excel in this field, Mr. Johnson Alabi, Dada

Mary Oluwanbo (my cousin), Mr. and Mrs. Ogunlola, Mr. and Mrs. Akinluyi, Mr. and Mrs.

5
Adetoyinbo, Dr. and Dr. (Mrs.) Ogunmola (Covenant Academy), Mr. Felix Dare Alabi, Akinluyi

Sola Mathew, (Femi, H.O.D. and other members of the 2013/14, 2014/15 and 2016/17 BFN

graduating set) and Jo Ann McCormick to mention a few. I also want to appreciate a special

mother, Mrs. Ruth Emume Babalola (a special mother, I love you so much) and my pastors;

Pastor Kayode Gabriel Abe, Pastor (Mrs.) Josephine Oluremi Abe, Pastor Taiwo Osadeyi, Pastor

Taiwo Ajobiewe, Pastor Ola Adejubee, Pastor Tunde Adelusi, Pastor Ilesanmi, Pastor Olu

Abiodun Sunday to mention a few and other members of RCCG Promised Land Parish, RCCG

Ekiti Province 4 Junior church, RCCG PSF, RCF EKSU. Also, my special appreciation goes to

my worthy mentor, Apostle Kola Mike-Bamigboye (K-Spirit) for his instructions and training.

I express my deepest thanks to Olaniyan Temitayo (T-Whizzy) for taking his useful time in

teaching and providing me a sound foundation as regards econometric analysis. I use this

medium to acknowledge his contribution greatly. In addition, my unreserved gratitude goes to

my roommates at one time or the other such as Akinrinade Mayowa, Ajewole Benjamin, Abe

Oreofe and other members of Cannanland and A and A Hostels. However, I do not want to

conclude this section without penning down my unflinching love and unreserved gratitude to

Babalola, Temitope Juliana who has been a darling, a dear and special friend who has stood with

me all the years through thick and thin all through my stay in EKSU. Finally, I appreciate all my

friends and classmates who have played one role or the other towards the successful completion

of this programme. The comprehensive list is enormous, but to mention a few they are; Abe,

Itunuoluwa Laughter, Abe Tomisin, Olabode Kemi, Moses Olayinka, Thelma, Aduloju Mayowa,

Omoloye Najeem (TeamNJB-a worthy friend), Madoti Halimat (Leemah), Lasekan Rebecca,

Olanrewaju John, Ajiteru Opeyemi and finally, the lovely Babalola, Temitope Juliana. Thanks to

everyone. I remain very grateful.

6
 ABSTRACT

The study examined the impact of monetary policy and its lag on economic growth in United

Kingdom, United States of America and Nigeria. The comparative study used Gross domestic

Product as the dependent variable and also used monetary policy rate, inflation rate, interest rate,

money supply and exchange rate as independent variables. The study used the Generalized

Method of Moments to examine the effect of lag, Auto Regressive Distributed Lag modeling

approach for long run analysis coupled with the Engle Granger causality test to reveal the

direction of causality. Hence, it was revealed that monetary policy lag has a negative effect on

economic growth in the three countries. Also, monetary policy was found to exert an

insignificant effect on economic growth in the United Kingdom and United State of America

while monetary policy was found to exert a significant effect on economic growth in Nigeria

through the channel of money supply. Meanwhile, the causality test revealed that economic

growth causes monetary policy in USA and Nigeria while both phenomena dictate each other‟s

tune in the UK. Hence, it was recommended that policy makers should focus on reducing the

existence of lags in the economy while monetary policy should not just be seen as a mere tool to

display the tyranny of monetary authorities in Nigeria but should be adopted a an effective tool

of response to put the economy on track after a careful observation of the economy has shown

that it is straying away from set targets and goals while the monetary policy should not be

aggressively used as the market force mechanisms should be allowed to determine prices and

phenomena within the economy. Also, monetary authorities should make use of money supply in

sharply correcting the economy in Nigeria in case of any perceived deviation from set targets as

this channel will bring about a substantial effect within the economy without so much delay to

ensure that Nigeria catches up with other industrialized economy of the world.

7
 Table of Contents

Front Page ………………………………………………………………………. i

Declaration ………………………………………….……………………….… ii

Certification ………………………………………….……………………….… iii

Dedication …………………………………………...……………………….… iv

Acknowledgements ………………………...……………………………….…. v

Abstract ……………………………………...………………………………….. vii

Table of Contents ………………………………………………………………. viii

List of Tables…………………………………………………………………… xii

List of Figures…………………………………………………………………… xv

CHAPTER ONE: INTRODUCTION

1.1 Background to the Study…………………………………………………. 1

1.2 Statement of the Research Problem………………………………............. 3

1.3 Objectives of the Study…………………….………………………............ 5

1.4 Research Questions…………………….................................................... 6

1.5 Research Hypothesis ……………………………….……………………… 6

1.6 Significance of the Study……………………………………….................. 7

1.7 Scope and Limitations of the Study……………………………………….. 8

1.8 Organisation of the Study…………………………………….................... 8

1.9 Operational Definition of Terms…….……………………………………… 9

CHAPTER TWO: LITERATURE REVIEW

2.1 Conceptual Literature …………………………………………………………. 11

8
2.1.1 Overview of Money…………………………………………………………… 11

2.1.2 Monetary Policy, its Lags and Economic Growth………………………..… 12

2.1.3 Monetary Policy Instruments……………………..…….…………………… 14

2.1.4 Monetary Policy: The Nigeria Experience…………………………………... 16

2.1.5 Monetary Policy: The Experience of the United States of America ….…… 18

2.1.6 Monetary Policy: The Experience of the United Kingdom ………………… 19

2.2 Theoretical Literature ……………………………………………..………... 20

2.2.1 McKinnon-Shaw Theory of Financial Liberalization…………………...... 20

2.2.2 Mundell-Fleming Model………...…..………………………………………. 21

2.2.3 Classical Quantity Theory of Money…………………………………………. 22

2.2.4 Keynesian Monetary Theory…..……………………………………………… 23

2.2.5 Friedman Restated Quantity Theory of Money………..…………………… 24

2.2.6 Endogenous Growth Theory…………………..………..…………………… 25

2.3 Review of Empirical Literature………………………………………………. 25

2.3.1 Empirical Evidence from Developed Countries………………..…………… 26

2.3.2 Empirical Review from Developing Countries……………………………… 31

2.3.3 Empirical Review from Nigeria………….…………………………………… 38

2.4 Gaps Identified………………..………….…………………………………… 44

CHAPTER THREE: RESEARCH METHOD

3.1 Theoretical Framework…………………………………………………………… 45

3.2 Model Specification……………………………………………………………… 45

3.3 Re-Statement of Hypothesis….…………………………………………………. 48

3.4 Estimation Techniques…………………………………………………………… 49

9
3.5 Econometric Tests……………………………………………………………. 50

3.5.1 Generalized Method of Moments (GMM)… ………………………………. 50

3.5.2 Unit Root Test (URT)………………………………………………………. 51

3.5.3 Co-Integration Test…………………………………………………………. 51

3.5.4 Standard Error Test ………………………………………………………… 51

3.5.5 Coefficient of Multiple Determination (R2)……………………………….. 52

3.5.6 Overall Significance of the Model (F-Test)……………………………….. 52

3.5.7 Durbin Watson Test (DW-Test)…………………………………………… 52

3.5.8 Auto Regressive Distributed Lag (ARDL) Model……………………….. 53

3.5.9 Model Stability and Diagnostic Test……………………………………. 54

3.5.10 Engle Granger Causality Test…………………………………................. 54

3.6 Sources of Data………….…………………………………………………… 55

3.7 Description of Variables …………………………………………………… 56

3.8 Theoretical Expectation ……………………………………………………. 57

3.9 Limitation of Data ………………………………………………………….. 58

CHAPTER FOUR: DATA ANALYSIS AND FINDINGS

4.1 Data Presentation………………………………………………………………….. 59

4.2 Generalized Method of Moments (GMM) …………………………………….. 59

4.3 Test for Significance of the Parameters in the Short Run (Probability Test)…. 64

4.4 Test for Stationarity of the Variables (Unit Root Test)…. ……………………. 67

4.5 Summary of Order of Co-Integration…………..……………………………… 74

4.6 ARDL Bounds Test Approach to Co-Integration…………………………… 76

4.7 Long Run Results………………………………………………………………. 79

10
4.8 Test for Statistical Significance of Parameters in the Long Run

(Probability Test)…………………………..………………………………….. 82

4.9 Diagnostic and Stability Tests………..………………………………………… 84

4.10 Interpretation of the Granger Causality Test………………………………. 92

4.11 Summary of the Research Findings………………………………………… 102

4.12 Implication of Research Findings………………………………………….. 104

CHAPTER FIVE: SUMMARY, CONCLUSION AND RECOMMENDATIONS

5.1 Summary….…………………………………………………………………… 108

5.2 Conclusion………………..…………………………………………………… 110

5.3 Recommendations ……………………………………………………………. 110

5.4 Areas of Further Research……………………………………………………. 112

References ………………………………………………………………………… 113

Appendix I…………………………………………………………………………. 119

Appendix II ……………………………………………………………………….. 125

11
 List of Tables

Table 2.1: Snapshot of Empirical Review from developed countries……. 30

Table 2.2: Snapshot of Empirical Review from developing countries…… 36

Table 2.3: Snapshot of Empirical Review from Nigeria…………………… 42

Table 4.1: GMM Result (Nigeria)…………………..……………………… 60

Table 4.2: GMM Result (USA)…………………..……………………… 61

Table 4.3: GMM Result (UK)…………………..……………………… 62

Table 4.4: Short Run Probability Test (Nigeria).…..…………………… 64

Table 4.5: Short Run Probability Test (USA).…..……………………… 65

Table 4.6: Short Run Probability Test (UK)…….…..…………………… 66

Table 4.7: Result of ADF Unit Root Test at Level for Nigeria …………… 69

Table 4.0: Result of ADF Unit Root Test at First Difference for Nigeria.… 70

Table 4.7: Result of ADF Unit Root Test at Level for USA……………… 71

Table 4.0: Result of ADF Unit Root Test at First Difference for USA.… 72

Table 4.0: Result of ADF Unit Root Test at Second Difference for USA.… 73

Table 4.0: Result of ADF Unit Root Test at Level for UK……………...… 73

Table 4.0: Result of ADF Unit Root Test at First Difference for UK…..… 74

Table 4.0: Summary of Order of Integration (Nigeria)……………………. 75

Table 4.0: Summary of Order of Integration (USA)………………………. 75

Table 4.0: Summary of Order of Integration (UK)…………………………. 76

Table 4.5: ARDL Co-Integration and Bound Test Result (Nigeria).……… 77

Table 4.5: ARDL Co-Integration and Bound Test Result (USA)….……… 78

Table 4.5: ARDL Co-Integration and Bound Test Result (UK)…………… 78

12
Table 4.8: Long Run Result of the Model (Nigeria)….……………………. 79

Table 4.8: Long Run Result of the Model (USA)…….……………………. 80

Table 4.8: Long Run Result of the Model (UK)……./.……………………. 81

Table 4.9: Long Run Probability Test (Nigeria)…….……………………………. 82

Table 4.9: Long Run Probability Test (USA)…….……………………………. 83

Table 4.9: Long Run Probability Test (UK)…….……………………………….. 84

Table 4.10: Result of the Breusch-Godfrey Serial Correlation LM Test (Nigeria).... 85

Table 4.10: Result of the Breusch-Godfrey Serial Correlation LM Test (USA).... 85

Table 4.10: Result of the Breusch-Godfrey Serial Correlation LM Test (UK)…... 86

Table 4.11 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (Nigeria)........ 87

Table 4.11 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (USA)........... 87

Table 4.11 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (UK)….......... 87

Table 4.13 Ramsey RESET Test Result (Nigeria).………………………….………… 88

Table 4.13 Ramsey RESET Test Result (USA).………………………….………… 89

Table 4.13 Ramsey RESET Test Result (UK).………………………….………… 89

Table 4.14 Causality between Monetary Policy Rate and

Gross Domestic Product (Nigeria)……………………………………… 93

Table 4.14 Causality between Inflation Rate and

Gross Domestic Product (Nigeria)……………………………………… 93

Table 4.14 Causality between Interest Rate and

Gross Domestic Product (Nigeria)……………………………………… 94

Table 4.14 Causality between Money Supply and

Gross Domestic Product (Nigeria)……………………………………… 94

13
Table 4.14 Causality between Exchange Rate and

Gross Domestic Product (Nigeria)……………………………………… 95

Table 4.14 Causality between Monetary Policy Rate and

Gross Domestic Product (USA)………………………………………… 96

Table 4.14 Causality between Inflation Rate and

Gross Domestic Product (USA)………………………………………… 96

Table 4.14 Causality between Interest Rate and

Gross Domestic Product (USA)…..…………………………………… 97

Table 4.14 Causality between Money Supply and

Gross Domestic Product (USA)..……………………………………… 97

Table 4.14 Causality between Exchange Rate and

Gross Domestic Product (USA)..……………………………………… 98

Table 4.14 Causality between Monetary Policy Rate and

Gross Domestic Product (UK)..………………………………………… 99

Table 4.14 Causality between Inflation Rate and

Gross Domestic Product (UK)..………………………………………… 99

Table 4.14 Causality between Interest Rate and

Gross Domestic Product (UK)……..…………………………………… 100

Table 4.14 Causality between Money Supply and

Gross Domestic Product (UK)…..……………………………………… 100

Table 4.14 Causality between Exchange Rate and

Gross Domestic Product (UK)…...……………………………………… 101

14
 List of Figures

Fig. 2.1: Transmission Mechanism through Interest Rate Channel……………… 14

Fig. 2.2: Instruments of Monetary Policy as it affects Economic Growth….…... 15

Fig. 4.1: Cumulative Sum of Squares (CUSUM) Test (Nigeria)………………… 90

Fig. 4.2: Cumulative Sum of Squares (CUSUM) Test (USA)………………..… 91

Fig. 4.3: Cumulative Sum of Squares (CUSUM) Test (U.K.)….……………….. 91

15
 CHAPTER ONE

INTRODUCTION

1.1 Background to the Study

Money is the bedrock of every economy. It has been considered as the major engine

behind the development and stasis of any economy based on the way its controlled. In line with

the above, the endogenous growth theory assumes that economic growth is spurred by factors

within the economy. Also, Fiador (2015) posited that the financial system of a country including

the rate of its development, rate of regulation and control has a major role to play in facilitating

economic development. Obviously, one of the major factors which engender growth is the nature

of the monetary policies as formulated by the monetary authority. Fiador (2015) corroborated

this truth as she averred that even in managing financial systems; the key policy tool is

undisputedly the monetary policy. Hence, the issue of cost of funds, volume of same, capital

accumulation, stability and interplay of forces within the economy are all results of the actions

and inactions of the monetary authorities through their policies. In support of the claim that

monetary policy is very important policy in the economy, Mishkin (2007) opined that the

monetary policy is a weapon that is needed for financial stability and health of any economy in

the world.

Specifically, monetary policy as explicated by Shaw (1973) refers to the conscious action

taken by monetary authorities to alter the quantity, availability or cost of money. Hence, Jain and

Khanna (2006) posited the economic development can be achieved through the adjustment of

money supply and its direction to the needs of development in the economy. They emphasized

that for economic development to take place, there is need to channel resources and funds to the

16
technological and infrastructural aspect of the society. The implementation of monetary policy as

delineated by Onderi (2014) boils down to the instruments, operating targets and the policy goals

of the policy which are the major components and factors to be considered in the implementation

of monetary policy. Onderi (2014) further considered price stability and economic growth as the

ultimate goals of monetary policy while other goals fall in as intermediate targets.

Although, monetary policy remains a major tool to navigate the economy through

economic vicissitudes, it can be disclosed that its improper use may as well hamper the growth

of any economy. For instance, Fiador (2015) considered contractionary monetary policy as a

policy that may lead to undesirable results in the economy. Hence, the weapon of monetary

policy is not only needed in an economy but the skill of use of it is likewise needed as the

economy grows. As a result, there is a call for emphasis on monetary authorities who adopt

various monetary policies in the economy. Although, the monetary authorities cannot affect

inflation and other set targets directly in the economy, they really influence this economic

phenomenon through some transmission mechanisms (Klasco, 2013). Correspondingly, Isedu

(2013) asserted that the effect which monetary policy will exert on the economy is majorly a

function of the ability of the monetary authorities to make precise and accurate assessment of the

timing and outcome of their policies and actions. Consequently, this need for accuracy in the

sphere of monetary policy has called for the consideration of lag in the process of formulating

monetary policies in every country.

In Nigeria, just like in every other country of the world, monetary policy has been

considered to be like a five pronged fork that is used to turn the economy around for better

premised on the concept of price stability, maintenance of external payment equilibrium,

promotion of employment, national output growth and sustainable economic development. The

17
economic development of the country has therefore be considered to rest on its monetary policy,

as a result, there have been reforms in the areas of infrastructure, power supply, electricity and

other sectors of the economy (Isedu, 2013). However, despite the universality of monetary policy

in various countries of the world, Nigeria has failed to make use of such universality has it has

focused mostly on the oil sector which has turned out to be a „Dutch disease‟ and its effect on the

economy. Also, the government overshoots its budget to a point that the funds which were

supposed to be used for infrastructural and social development have never been employed while

monetary authorities have been forced to increase interest rates because of inflationary pressures

(Isedu, 2013) coupled with the presence of information lopsidedness as a result of the state of the

economy necessitating the unavoidable presence of monetary policy lag. As a result, it is against

this backdrop that this study seeks to examine the impact of monetary policy and its lag on

economic growth coupled with a comparative analysis with two other countries of the world viz;

the United States of America and the United Kingdom.

1.2 Statement of Research Problem

Economic growth responds significantly to monetary policy all over the world and

Nigeria is not an exception. However, there exists a plethora of contribution in literature as

regards this subject matter (Kutu, Nzimande & Msomi, 2017; Borio & Gambacorta, 2017;

Drama, 2017; Inam & Ime, 2017; Najaf, 2017; Chan, 2016; Anowor & Okorie, 2016; Wong &

Chong, 2015; ThankGod & Karimo, 2014). These studies range from those carried out on

industrially, economically or technologically developed countries to those carried out on

developing countries. However, due to the economic crisis that has rocked the world, there is

need to revisit the influence of monetary policy on economic growth across the world. As a

18
result, this study carried out a comparative analysis of the effect of monetary policy on economic

growth.

Furthermore, considering the conflicting result in literature as regards monetary policy

whereby the financial liberalization school of thought assumes that financial liberalization

engenders economic growth while the financial repression school of thought assumes that

repression improves growth. In the same manner, Kutu, Nzimade and Msomi (2017), Chan

(2016) and Wong and Chong (2015) discovered that monetary policy has significant effect on

economic growth while Drama (2017), Tang (2009) and Bokreta and Benanaya (2016)

discovered otherwise. Also, AbuDalu, Elsadig, Almasaled and Abuelgasim (2014), Irfan and

Amen (2011) and Anowor and Okorie (2016) discovered that monetary policy has positive effect

on economic growth while Srithilat and Sun (2017) and Obadeyi Okhiria and Afolabi (2016)

discovered otherwise and interestingly, Douanla (2014) discovered both effects on economic

growth. Hence, given this mixed results it is necessary to inquire into the effect of monetary

policy on economic growth to ascertain the real fact.

Also, from a cross examination of several contributions in literature, more knowledge

gaps were identified which this present study intends to fill. Chan (2016) in a research study on

the subject matter in China failed to examine the causal relationship between the variables and

also did not take into consideration the effect of lag in the economy. Also, Asghar and Hussain

(2014) employed the short run econometric technique to examine the subject matter in Pakistan.

However, a close look at the study revealed that it relied on a short span of study, failed to reveal

the causal relationship and consider the lags in the economy. In like manner, Liu (2012) worked

on related subject matter in Australia but also suffered from the setbacks of the above

contributions to literature. Furthermore, from the Nigerian standpoint, most studies (Inam & Ime,

19
2017; Anowor & Okorie, 2016; Akinjare-Victoria, Babajide and Okafor, 2016; Obadeyi, Okhiria

& Afolabi, 2016) have failed to establish the direction of causality between the variables and

were limited to the Error Correction Modeling technique. Also, these studies failed to consider

the effect of monetary policy lag on economic growth and obviously failed to consider a

comparative analysis with other countries of the world on the subject matter. Hence, this study

investigates comparatively the effect of monetary policy and its lag on economic growth of

Nigeria, United States of America (USA) and the United Kingdom (UK) through the use of the

Auto Regressive Distributed Lag Modeling (ARDL) and the Generalized Method of Moments

(GMM) techniques respectively coupled with the use of the Engle Granger Causality technique

to examine the causal relationship between monetary policy and economic growth between 1986

and 2016. It is however imperative to note that the Generalized Method of Moments (GMM)

technique is employed to assess the effect of monetary policy lag on economic growth.

1.3 Objectives of the study

The broad objective of this study is to investigate the impact of monetary policy on economic

growth in selected countries while the specific objectives are to:

i. examine the effect of monetary policy lag on economic growth.

ii. provide evidence on the causal relationship between monetary policy indices and

economic growth.

iii. assess the comparative country-specific effect of monetary policy on economic

growth.

20
1.4 Research Questions

The subject of investigating the effect of monetary policy on economic growth has been

one of the most debated in finance and development literatures, as the conduct of further research

continues to generate controversies and mixed results, this study seeks to contribute to the

existing debate by investigating empirically whether or not monetary policy has any effect on

economic growth in the selected countries as well as direction of causality in the study.

In the light of the objectives of the study, an attempt will be made to provide answers to the

following pertinent questions:

i. what effect does monetary policy lag have on economic growth in the long run?

ii. What are the monetary policy indices that cause economic growth in Nigeria?

iii. To what extent does country fine tune the effect of monetary policy on economic

growth?

1.5 Research Hypotheses

Oloyede (2002) defined hypothesis as a tentative, testable and verifiable statement about

the relationship that subsists between two or more economic variables. Also, Encarta dictionary

also explained hypothesis to be a tentative explanation for a phenomenon used as a basis for

further investigation, an assumption needing investigation.

Therefore, for the the purpose of this research, the hypotheses to be tested will be

formulated in it null form (H0). The Null hypothesis is usually stated in negative form. The

hypotheses that are of necessity to be tested in this study are stated in nulls:-

i. Monetary policy lag has no effect on the economic growth in the long run.

21
 ii. There exists no causal relationship between monetary policy indices and economic

growth.

iii. Country does not fine tune the effect of monetary policy on economic growth.

From the above speculations, the hypothesis for the study can therefore be formulated below as:

H0: Monetary Policy has no significant impact on economic growth of the selected countries.

1.6 Significance of the Research

Literature is replete with preponderance of empirical studies (Kutu, Nzimande & Msomi, 2017;

Borio & Gambacorta, 2017; Drama, 2017; Inam & Ime, 2017; Najaf, 2017; Chan, 2016; Anowor

& Okorie, 2016; Wong & Chong, 2015; ThankGod & Karimo, 2014) relating to the impact of

monetary policy on economic growth. This study is conducted to add existing body of

knowledge in literature on this subject matter.

However, previous studies relating to Nigeria and beyond have suffered from the use of

limited predictors in explaining monetary policy coupled with lack of update and current

relevance. These studies have also failed to establish the causal relationship that exists between

monetary policy and economic growth while they have also failed to adopt a comparative

analysis with other countries of the world. However, apart from the utilisation of the Ordinary

Least Square Technique, multivariate co-integration technique and the Error Correction

Modelling, the study also employed the Engle Granger Causality test to determine the direction

of causality.

Moreso, this study is expected to provide the monetary authorities and policymakers,

investors, observing public and economic experts with astounding insights. As a result, policy

recommendations will be developed to equip various economic entities in Nigeria with the

22
acumen needed to spur the growth of the Nigerian economy through the right use of the

monetary policy.

In addition, this study will be a reference for further research on the same or related topic

among academicians and scholars. Finally, to the ordinary reader, this study will serve as an

immense discovery and valuable store of knowledge.

1.7 Scope and Limitations of the Study

The purpose of this study is to investigate the impact of monetary policy on economic

growth of Nigeria. It is necessary to note that the data that will be analysed in the current study is

the data for the economy of the selected countries viz; Nigeria, United States of America and

United Kingdom obtained from secondary sources. The scope for this study spanned from 1986

to 2016, as the time series annual data for the period will be used. The justification for the scope

is not just because the researcher wanted to pick 1986 as the starting year, but because it marked

the beginning of the Structural Adjustment Programme, the advent of financial liberalisation

which is one of the major monetary policies in Nigeria.

The major limitation in this regard is the aspect of data collection. Most of the data

collected through print journals and publications required different management. Also, there

exists a dearth of literature on the subject matter especially recent contributions from developed

countries of the world. Also, financial constraint was also another major factor of limitation.

Nonetheless, irrespective of the limitations, the study was be conducted in a manner that the

objectives of the study was not jeopardised.

1.8 Organisation of the Study

This study is segmented into five comprehensive chapters. Chapter one provides

background and the general introduction of the subject matter. The chapter states the research

23
problem, objectives of the study, research questions as well as the hypothesis which are intended

to be tested in the course of the study, the significance of the study, scope and limitations and

lastly the outline for the study. Chapter two entails the conceptual issues, theoretical

reinforcement and groundwork, alongside various reviews of empirical studies on the impact of

monetary policy on economic growth.

Furthermore, the methodology, where the model to be adopted in the study is specified,

the method of analysis as well as the estimation procedures form the pedestal of chapter three.

Chapter four is exclusively reserved for analysis of data gathered for the purpose of study and

then the interpretation of findings while chapter five which is the last segment presents the case

for a befitting summary, remarkable conclusion, and relevant policy recommendations based on

the strength of the findings emanating from the current study.

1.9 Operational Definition of Terms

Monetary Policy

This refers to the policy of the government aimed at controlling and regulating the

volume, direction and the cost of money in the economy.

Financial Liberalisation

From the layman‟s point of view, financial liberalisation is the removal or relaxing of

restrictions imposed on the domestic financial market by the government. Financial liberalisation

involves deregulation of the foreign sector capital account and the stock market.

Structural Adjustment Programme

This is a term that refers to the deregulation of the financial sector of the Nigerian

economy. The programme was initiated to remove the regulation of the financial sector by the

government either fully or partially and allow the interplay of the market forces boost economic

24
growth. Measures adopted during the programme include the deregulation of interest rate,

exchange rates and sectoral allocation of credit.

Financial System

The financial system refers to the network of financial institutions, markets and entities

existing in an economy at a particular time responsible for intermediation and efficient allocation

of financial resources.

Expansionary Monetary Policy

This is a monetary policy targeted at increasing money supply and injecting capital into

the economy in a bid spur economic growth.

Contractionary Monetary Policy

This is a policy targeted at reducing government spending and monetary supply or

expansion in a bid to combat inflation.

Monetary Policy Lag

This refers to the length of time that elapses between when there is need for an action,

when the action is taken and when the effect is felt on the economy.

25
 CHAPTER TWO

LITERATURE REVIEW

2.1 Conceptual Literature

This chapter lays to bare the existing theories and studies relating to monetary policy and

economic growth. The nub of this study is to equip the reader with the adequate knowledge

apropos how monetary policy affects economic growth. Hence, at the end of this chapter the

reader will have a better understanding of conceptual issues relating to the subject matter. This

chapter will be reviewing several contributions in literature on the same or related topics

conducted by scholars of finance and development within and outside Nigeria. This study centers

its attention on the analysis of monetary policy in a country like Nigeria and its immense benefit

to the economy.

2.1.1 Overview of Money

Money has been found to have so many uses and definitions that it is very difficult to

conclude on a single definition and use of money. In line with the above, Scitovsky (1940)

posited that money is not a phenomenon that is easily defined in that it is a unit of account, store

of value and also a medium of exchange. Hicks (1935) as relayed by Smitha (2010) considered

money to be anything that money can do while Crowther (1948) defined same as any material

that is generally acceptable as a medium of exchange and simultaneously acts as a measure and

store of value. As a result, it can be extrapolated that money is used in every economy but the

major difference is the material that stands for money in the various economies of the world.

26
 Correspondingly, Smitha (2010) considers money as the oil to the wheels of every

economy in the world. He further considered the functions of money to be into categories viz;

the primary, secondary and contingent functions. Primarily, Smitha (2010) considered money to

be a medium of exchange of goods and services and as a unit of value, that is, a yardstick for

measuring value in the economy. He moved an inch further by considering money as a

phenomenon that performs some secondary functions which includes being a standard for

deferred payments or postponed payments. Smitha (2010) explicated the contingent function of

money to include its liquid nature which enables it to be readily spent at any time and its use as a

measure of income especially national income. However, premised on these opinions, it can be

inferred that the economy thrives on money; as a result, any change in money supply can greatly

affect the rate of economic growth. This singular reason has necessitated the need for policy

makers to pitch their tents around policies relating to the amount, direction and flow of money in

the economy otherwise known as monetary policy.

2.1.2 Monetary Policy, its Lags and Economic Growth

Monetary policy refers to the policies and measures adopted by the government of a

country to control the value, availability, direction, cost and supply of money in an economy in a

bid to achieve set objectives. However, Smitha (2010) opined that monetary policy relates to

non-monetary measures that are targeted at influencing the monetary environment such as

regulation of prices, wages and income. Johnson (1978) considered monetary policy as the

policy employed by the monetary authority of a nation in a bid to control the supply of money in

order to achieve the macroeconomic objectives of the government. In the same manner, Einzing

(1964) opined that monetary policy embraces all the decisions of the government that has

sustaining effect on the monetary system either they are monetary decisions or not.

27
 Correspondingly, monetary policy has been considered as the Sulaiman (2006) as

conscious effort by monetary authorities in a country to control the stock of money and the

creation of credit through the use of quantitative and qualitative tools. Also, Cambazoglu and

Karaalp (2012) posited that the transmission mechanism of monetary policy, that is the channels

through which monetary policy affect economic growth can be split into the neo-classical

channel and the non-neoclassical channel. They averred that the neo-classical channel relates to

the traditional interest rate channels, asset price channel and the exchange rate channel.

Meanwhile, the non-neoclassical channel involves majorly the credit or money supply channel.

Furthermore, the objectives of monetary policy involves ensuring price stability,

economic growth, exchange rate stability, full employment and the balance of payments

equilibrium. However, the achievement of these objectives has called for the consideration of

lags in the implementation of monetary policy. As a result, Sulaiman (2006) considered time lag

as the major lag affecting monetary policy in the economy. Time lag is considered as the length

of time that exists between when there is need for a monetary policy action and when the action

is taken. It is also the length of time between when an action is taken and when the effect of the

action is felt in the economy. According to Friedman (1961)1 monetary policy effect is

substantially felt in the economy only after a length of lag, as a result, he recommended that

policy makers should forecast the existence of lags in the process of making policies to reduce

the problem of instability in the country. Sulaiman (2006) further delineated time lags to involve

inside and outside lag whereby the inside lag is the length of time between the need for an action

and when it taken while outside lag is known as the time that lapses between when an action is

1
The proposition of Friedman was refuted by Culbertson (1960) in his paper “Friedman on the
lag in effect of Monetary Policy”

28
taken and when the effect is substantially felt in the economy. However, irrespective of the kind

of lag, monetary policy lags have been found to exert negative effect on economic growth

(Sulaiman, 2006).

Fig. 2.1: Transmission mechanism through Interest rate channel

Source: Bazina (2012)

2.1.3 Monetary Policy Instruments

These are techniques used by monetary authorities to control, regulate and direct the

volume, flow and direction of money in the economy. In developing countries like Nigeria,

monetary policy techniques includes the Open Market Operation (OMO), Moral Suasion, Legal

Reserve Ratio, discount rate and special deposit (Apinran, 2015).

Specifically, the Open Market Operation (OMO) technique refers to an indirect monetary

policy technique used in regulating the volume of money in the economy. It is designed to

promote non-inflationary economic growth as it involves the monetary authorities selling and

buying securities in the market in a bid to reduce and pomp cash into the economy respectively,

however, the effect of OMO is not very instantaneous on the economy if it is not developed.

Also, the discount rate of the monetary authority which refers to the rate at which credit is

29
supplied to commercial bank is another major tool in controlling the economy. Monetary

authorities spur the rate of liquidity in the economy by reducing the discount rate necessitating

the reduction of the rate at which banks give out loans and as a result, volume of money supply is

increased and vice versa (Apinran, 2015). Furthermore, the reserve requirement ratio refers to the

part of the funds of commercial bank that must be kept with the monetary authorities, in a

situation whereby monetary authorities wants to reduce or mop up excess funds in the economy,

the ratio is increased while in a bid to increase money supply, the ratio is reduced.

Also, moral suasion refers to the gentle appeal and persuasion by the monetary authorities

of a country to follow a particular laid down guideline of the government. This is done because

the monetary authorities assume that the banks should know how to carry out their activities

within the whims and caprices of their directives. Sulaiman (2006) posited that for moral suasion

to work very well in the economy, the financial institutions must have close affinity with the

monetary authorities in the economy. Meanwhile, special deposit refers to the directive of the

monetary authorities to the deposit money banks to deposit a particular sum of credit with them

in a bid to mop up excess funds in the economy and reduce cash with the banks to curb inflation.

Instruments:

- Open Market
Operations
Monetary Money
- Legal Reserve Economic
Policy Supply
Requirements Growth
- Discount Rate
- Moral Suasion

Fig. 2.2: Instruments of Monetary Policy as it affects economic growth

Source: Author‟s Design (2017)

30
2.1.4 Monetary Policy: The Nigeria Experience

The monetary authority in Nigeria is the Central Bank of Nigeria and just like in every

other country of the world; it has considered growth as one of its monetary policy objectives

(Isedu, 2013). Over the years, the monetary policy of the Central Bank of Nigeria hinges on four

major pillars viz; price stability, real output growth, reduction in the unemployment rate,

maintaining healthy balance of payments and increased savings and credit flow to the major

sectors of the economy. At the early stage of the economy, the financial system was driven by

the concept of the “cheap money policy” which was accomplished through the creation of local

currency, the introduction of the ,money and capital market and keeping interest rates low for

target sectors of the economy (Sanusi, 2002). Right from inception till 1979, the CBN adopted

the quantitative and qualitative monetary measures in regulating the economy as a whole.

According to Isedu (2013), the quantitative measures relates to the measures that handle the

quantity, volume and price of credit in the economy. One major instrument adopted as a

quantitative technique is the Open Market Operations (OMO) which involves the buying and

selling of debt instruments by the government. On the other hand, the qualitative measures

relates to those which determines the direction and distribution of credit like moral suasion.

Although, prior to 1991, the CBN did not adopt the OMO technique due to the under

developed nature of the economic system, direct monetary regulations were used as monetary

measures as the forces of demand and supply were not allowed to determine financial and

economic prices and rates. In 1976, stabilization securities were first introduced into the

economy in a bid to ensure that banks accept more of deposits as instruments of monetary

management. However, as at this period, the most effective measure adopted then was the credit

guidelines as introduced in 1964 (Isedu, 2013). However, until 1986 which marked the

31
introduction of the Structural Adjustment Programme (SAP) era, the Nigerian monetary

authorities adopted the direct monetary control system whereby all rates in the economy

including the interest rate which happened to be the major instrument of monetary policy was

strictly determined by the authorities. Prior to the SAP era, interest rate was fixed at a low

pedestal which led to the provision of cheap funds to the government and private individuals

(Isedu, 2013). This expansionary monetary policy resulted to an increase in credit base.

Consequently, an end came to the regime of the direct monetary control system with the

advent of the Structural adjustment programme in 1986 which was marked with the removal of

credit controls and other strict monetary regulations as the forces of demand and supply in the

market were allowed to determine the prices and rates in the economy. The SAP era was marked

by the abolition of credit controls, motivation of the non-oil sectors, liberalization of external

trade and payments system, reduction in administrative control and interference in the monetary

regulation of the country and obviously, restructuring of public expenditure the deregulation of

the economy as a whole (Isedu, 2013).

Moreover, succeeding the SAP period starting from1994 till date was marked with was is

termed as the „guided deregulation monetary system‟ in a bid to curtail the activities of players in

the economy. This was due to the side effects of the liberalization or the implementation of the

SAP programme in 1986. The effects such as inflation, proliferation of banks, fiscal deficits were

discovered; the major eyesore was the rate of inflation that rose from 5.7% in 1986 to 57% in

1994. As a result, investment was discouraged, exportation reduced significantly and the banking

system began a journey into serious banking distress. As a result, the monetary authorities

introduced some checks and balances to curtail the activities within the economic system.

Recently, the cashless policy was introduced in 2012 while bank consolidation took place in

32
2004 which led to the strengthening of the financial system. Hitherto, monetary authorities in

Nigeria has handled the issue of monetary policy well and has kept the economy in good shape

as money supply increased to 21,607 billion naira in 2016 while inflation and interest rates

remained at 18% and 5% respectively as at 2016.

2.1.5 Monetary Policy: The Experience of the United States of America (USA)

Due to the creation of the bank of England in 1694 as a monetary authority, the concept

of monetary policy became an issue to be adopted by developed countries of the world. The

major aim of monetary authorities in developed nations of the world relates to maintenance of

gold standard which requires monthly adjustments of interest rates. During the period which

lasted between 1870 and 1920, developed countries of the world established monetary authorities

around the world and the last one was the Federal Reserve system established in the USA in

1913 (FRB, 2003). In the USA, monetary authority evolved based on the banking system of the

country, the act establishing the Bank of North America in Philadelphia was established in 1781,

this bank was granted monopoly to issue bills of credit as currency at the national level and to act

as the sole fiscal and monetary agent for the government. However, in 1791, the First Bank of

the USA was established to succeed the Bank of North America as the major monetary authority

in the USA. In 1816, the second bank was established to succeed the First Bank as monetary

authority in the country.

Consequently, in between 1837 and 1863, the monetary authorities in the USA relaxed

the banking regulations and allowed the free banking concept in the country. However, as a

result of the free banking era, many banks became unstable and in 1863, the national banking act

which reinforced monetary authority in the USA again was established. Finally, in 1913, the

33
Federal Reserve System came to stay in response to recession in the economy. The Federal

Reserve System evolved to become a formidable monetary authority in the country.

2.1.6 Monetary Policy: The Experience of the United Kingdom

The monetary authority in the UK is the Bank of England which was established in 1694

although it was funded by private shareholders until it was nationalized in 1946 (BEA, 1946). In

1998, the Bank became independent and wholly owned by the treasury solicitor (BOE, 2013)

with independence of setting monetary policies. The bank has monopoly of issuing bank notes in

England and Wales but regulates issuing of notes in Scotland and Northern Ireland. In 1931, the

country Britain which previously maintained the gold standard stopped the usage of it. In 1945,

the bank adopted the policy of „easy money‟ and low interest rates (the expansionary monetary

policy) in a bid to support aggregate demand (Fforde, 1992). However, it fixed exchange rates at

same time, Furthermore, in 1981; the banks abolished the reserve requirement in the economy in

a bid to relax stringent monetary conditions.

In a bid to support the operations of the bank, the Monetary Policy Committee (MPC)

was established in 1998 with the sole responsibility of fixing interest rates in commensuration

with the rate of inflation. The aim of the monetary authority in the UK dwells so much on

maintaining stable prices and confidence in the currency. Furthermore, it acts as the banker of

the government and manages the gold reserves of the country. Also, in 2011, the financial policy

committee was established to ensure financial stability in the country necessitating the macro

prudential regulation of the banks of the UK. Ever since then, the monetary authorities and their

committees have endeavored to be their best at ensuring monetary stability in the United

Kingdom.

34
2.2 Theoretical Literature

There are several theories in economics and finance literature that offers theoretical

explanation on the link between monetary policy and economic growth. These theories will be

discussed comprehensively as obtained below:

2.2.1 McKinnon-Shaw Theory of Financial Liberalization

The financial liberalization theory started with the opinions of McKinnon (1973) and

Shaw (1973)2 while building on the work of Schumpeter (1911). The theory emphasized on

financial liberalization by considering the adverse effect of financial repression on economic

growth. They posited that most developing countries are in such state because they are

financially repressed, and as such, this repression has caused indiscriminate alteration and

fluctuations of financial prices such as interest and foreign exchange rate.

Correspondingly, Shaw (1973) posited that liberalization of the economy leads to the

increasing of the private savings to income in the economy. Hence, Isedu (2013) averred that

financial liberalization will lead to significant economic benefits through the mechanisms of a

more effective domestic savings mobilization, financial deepening and proper allocation of

resources. Furthermore, the proponents of this theory assumed that the existence of repression or

fixing of rates and allocation of credit by monetary authorities through their polices has

destroyed the economies of developing countries by leading to a downward trend in savings and

led to channeling of funds and encouragement of investment in unproductive sector of the

2
Accroding to Gemech and Struthers (2003) “The McKinnon-Shaw Hypothesis is now more than
thirty years old and throughout this period several empirical literature has been published to
verify its reality, though it first focused on financial repression, it now focuses on financial
liberalization.”

35
economy. Mckinnon (1973) and Shaw (1973) challenged the classical theory and propounded the

theory against financial repression assuming that interest rate should be determined by the

market and not by government. Furthermore, liberalization encourages competition and

innovation in the market, as a result, McKinnon (1973) assumes that such competition in the

financial market will raise interest rates on deposits and lead to a high savings rate and also lead

to financial innovation and import of proper managerial techniques in the economy.

2.2.2 Mundell-Fleming Model

The model has formulated by Mundell and Fleming (1963) as cited by Isedu (2013)

considers the IS-LM model by Hicks (1937) as relayed by Iseidu (2013) as its bedrock for its

assumptions and postulations. Due to the inability of the Keynesian and Classical models to

address the combined effect of unemployment and inflation rather than the inverse relationship

as considered by the Keynesian theory.

Specifically, considering the need for a country to achieve both a strong internal as well

as external balance, the proponents of this theory as relayed by Isiedu (2013) assumed that there

is a need to make use of both the monetary and fiscal policy efficiently and simultaneously.

Mundell-Fleming (1963) as quoted by Iseidu (2013) considered internal balance as domestic

balance where there is full employment and stability of prices in the economy while external

balance was seen as the equilibrium of the balance of payments. The advocates of this theory

therefore posited that in a bid to achieve internal balance in the economy, there is need to achieve

a tradeoff between inflation and unemployment and this is where the essence of monetary and

fiscal policies comes into play. However, in a bid to achieve external balance, Mundell-Fleming

(1963) as alluded to by Iseidu (2013) assumed that there is a need to achieve equilibrium

36
between exports and imports. They further assumed that the expansionary monetary policy can

be achieved through the process of lowering the interest rate in the economy as it will spur the

level of income and employment but it will also lead to increase in imports. Moreso, they

assumed that the contractionary monetary policy can be achieved by increasing the rate of

interest which will lead to a reduction in the level of income, investment, imports and

employment.

2.2.3 Classical Quantity Theory of Money

The quantity theory of money was first developed by Jean Bodin in 1958 and was later

refined by Irving Fisher in 1911 (Jhingan, 1997). The classical economists assumed that money

is used only for transactionary purposes and as such are only used for exchange of goods and

services. As a result, the quantity theory of money assumes that there is a strong connection

between money supply and price level Ceteris Paribus, i.e. if all things are equal. The widely

known equation in the theory is the famous Fisher‟s equation of exchange as modeled by Fisher

(1911) which is identified as:

MV = PT……………………………………………………………………… (2.1)

In the equation, „M‟ is considered to represent the quantity of money in circulation, „V‟

represents the number of times a unit of money is used in transaction per unit of time (velocity),

„P‟ represents the weighted average of all individual price, hence P = , meanwhile, „T‟ refers

to the sum and value of all the transactions per goods and services per unit of time.

However, the Fisher‟s equation of exchange was criticized based on the ground that

velocity of money is not constant and it emphasized a lot on the volume of transactions money is

37
used for rather than output which is the amount of goods and services exchange for the money

itself. As a result of these criticisms, the equation was transformed as T which is the volume of

transactions money is used for was substituted for Y which is the total amount of goods and

services (Dalhatu, 2012).

MV = PY……………………………………………………………………… (2.2)

Therefore, as a result of the introduction of Y into the equation, the theory shows that the

monetary side of the economy is linked with the real sector of the economy as Y stands for real

output. Hence, the postulation of the monetarist theory that a rise in money supply has no effect

on real economy but only on the monetary side and the prices at which goods are exchanged is

upheld (Onderi, 2014). In same manner, when money supply is reduced, the spending capacity of

the people is reduced and this leads to a fall in prices (Dwivedi, 2010). However, the model has

been criticized on the grounds that it did not sufficiently explain the general price movement and

that the postulation that there is direct proportion in change in quantity of money and price level

is non-existent today as there are other exogenous factors outside the model that affects changes

in quantity of money and price level like infrastructural and political factors (Dalhatu, 2012).

2.2.4 Keynesian Monetary Theory

This theory as propounded by Keynes (1936) considered the essence of money to

transcend just exchange of goods and services (transactionary motive). He asserted that money

can also be used for precautionary and speculative purposes that is, guarding against unforeseen

circumstances and for investment purposes. Keynes (1936) connects the demand for money to

the fluctuations in interest rates. He further emphasized his point through the use of bonds price

as example. He assumed that people hold money to buy bonds in the future expecting the prices

38
to go down. According to Dwivedi (2010), the theory clearly spells out three motives for holding

money which includes the precautionary, transactionary and speculative motives of holding

money.

Furthermore, Keynes (1936) considered the demand for money (Md) be a function of the

transactionary (which includes precautionary i.e MT) and speculative demand for money (MSP).

Hence, he formulated another equation identified as:

Md = MT + MSP ……………………………………………………………….(2.3)

Keynes (1936) assumed that the assumption of the constant velocity of money as propounded by

the classical economists is a mere fallacy and that nominal interest rate is a more important factor

that affects the velocity of money in an economy. As a result, he posited that demand for money

is negatively related to interest rate which is a major divergence from the classical quantity of

money.

2.2.5 Friedman Restated Quantity Theory of Money

This theory as postulated by Friedman (1956) considered money as an asset and that it

competes with other assets in the economy such as bonds, stock and physical goods in a bid to

gain a place in their portfolio. Freidman (1956) further asserted that the marginal utility of

monetary services reduces as the quantity held increases.

Furthermore, he considered the demand of money to be a function of real permanent

income, rate of return on bonds, rate of return on equities, rate of return on money itself and the

anticipated inflation rate. He states the equation in the manner below:

= f ( Yp , Rb , Re , Rm , ) …………………………………………………….(2.4)

39
Where, refers to the demand for money, Yp refers to real permanent income, Rb refers to the

rate of return on bond, Re refers to the rate of return on equities, Rm refers to the rate of return on

money and refers to the expected inflation rate. Friedman further averred that the demand for

money is essentially stable and does not depend on interest rates as posited by Keynes (1936).

2.2.6 Endogenous Growth Theory

The theory as advanced by Romer (1986) and Lucas (1988) assumes that the factors

leading to the growth of an economy are endogenous in nature which points to the fact that the

actions and inactions of monetary authorities through their various monetary policies within the

country influences economic growth. Froyen (2009) corroborated the theory by positing that an

increase in capital investment spurs economic growth; meanwhile, technological improvement

should be considered an endogenous factor rather than an exogenous factor. Furthermore, the

technological improvement is a function of research within the country. Finally, investment in

human capital is essential to the growth of any economy. Arrow (1962) and Rebelo (1991)

assumed that the endogenous nature of the growth process is revealed in the connection between

investment and technology diffusion. It was assumed that technology diffusion and knowledge

diffusion within the economy is also a major contributing factor to economic growth as firms

grow based on innovation which is a function of the interaction of knowledge and technology.

This model assumes that output increases with an increase in capital, labour and technology with

the assumption that these factors can exhibit either increasing or constant returns to scale.

2.3 Review of Empirical Literature

There are vast amount of vivid research literature on the impact of monetary policy on

economic growth across the world. This study seeks to provide empirical evidence on the subject

40
matter. However, several studies have been conducted in the past in line with the focus of this

research study. Hence, it is imperative to strategically review previous research so as to ascertain

whether evidence provided by this study would contribute significantly to existing body of

empirical literature.

Based on the theoretical literature, it is equally important to consider empirical studies

that reveal how monetary policy and its lag affect economic growth. The empirical review is

presented in three sections. The first section considers empirical review of studies carried out on

developed or industrially advanced countries of the world; the second section embraces empirical

review of studies carried out on developing or emerging countries of the world while the third

section focuses on empirical review on studies carried out on Nigeria which is the home country

of the study.

2.3.1 Empirical Evidence from Developed Countries

Some brilliant attempts have also been made by researchers in developed countries of the

world to investigate the impact of monetary policy on economic growth.

Kutu, Nzimande and Msomi (2017) studied the effect of monetary policy on growth of

industrial sector in China between 1994 and 2013. The study used industrial output production as

the dependent variable and also used interest rate, money supply, exchange rate and inflation rate

as independent variables coupled with the use of Auto Regressive Distributed Lag (ARDL)

modeling technique; it was revealed that monetary policy has significant impact on economic

growth. Hence, it was suggested that interest rate should be stabilized to boost industrial

production in the land.

41
 Borio and Gambacorta (2017) studied the effect of monetary policy on bank lending in 14

developed countries between 1995 and 2014. The study used annual growth rate of loans as

dependent variable and also used lending rate, inflation rate and price growth as independent

variables coupled with the use of the Generalized Methods of Moments, it was revealed that

monetary policy has negative effect on bank lending. Hence, the setting of rates at low level was

recommended to boost bank lending and growth of the economy.

Chan (2016) studied the effect of money supply and inflation rate on economic growth in

China between 1999 and 2015. The study used Gross domestic product as the dependent variable

and also used money supply and inflation rate as the independent variables coupled with the use

of the Vector Error Correction Modeling technique, it was revealed that money supply and

inflation rate as monetary policy variables have significant effect on economic growth. Hence, it

was suggested that a cautious regulation should be implemented to boost economic growth.

Mushtaq and Siddiqui (2016) studied the effect of interest rate as a monetary policy

variable on bank deposits in 46 countries across the world between 1999 and 2014. The study

used bank deposits as the dependent variable and also used interest rate as the independent

variable coupled with the use of the Panel Auto Regressive Distributed Lag modeling technique;

it was revealed that interest rate has positive effect on bank deposit. Hence, it was recommended

that depositors‟ behaviour should be considered while making policies.

Wong and Chong (2015) studied the effect of monetary policy on economic growth of

228 countries across the world between 1974 and 2009. The study used Gross domestic product

as the dependent variable and also used inflation, exchange rate and capital openness as

independent variables coupled with the use of the ordinary least square technique, it was

42
revealed that monetary policy has significant effect on economic growth. Hence, it was

recommended that inflation and exchange rates should be the target of monetary authorities in

the process of controlling the economy.

AbuDalu, Elsadig, Almasaled and Abuelgasim (2014) studied the effect of exchange rate

on economic growth of 5 Asian countries between 1991 and 2006. The study used Gross

domestic product as the dependent variable and also used exchange rate, interest rate, money

supply, inflation rate, terms of trade and net foreign asset as independent variables coupled with

the use of the Auto Regressive Distributed Lag modeling technique, it was revealed that

exchange rate and other monetary policy variables have positive effect on economic growth.

Hence, it was suggested that policy makers should integrate their short and long term policy

actions together.

Douanla (2014) studied the effect of monetary policy on economic growth in 14 countries

in the Franc zone between 1985 and 2012. The study used Gross Domestic Product as the

dependent variable and also used money supply, total reserves, domestic credit, trade, investment

and inflation as independent variables coupled with the use of Generalized Methods of Moments;

it was revealed that monetary policy has both positive and negative effect on economic growth.

Hence, it was recommended that funds should be allocated to projects with highest social return.

Liu (2012) studied the effect of monetary policy and bank size on bank lending in

Australia between 2004 and 2010. The study used bank loans as dependent variable and also

used cash rate, liquidity risk, deposit rate and confidence as independent variables coupled with

the use of descriptive statistics and regression technique; it was revealed that monetary policy

43
has significant effect on bank lending. Hence, it was recommended that bank size should be

considered in monetary control.

Ridhwan, Henri, Nijkamp and Piet (2010) studied the impact of monetary policy on

economic growth in US and Europe. The study used stock market capitalization, inflation rate,

size and exports as independent variables coupled with the use of meta-analysis technique as

regards OLS and VAR regression technique; it was revealed that monetary policy has significant

effect on economic growth.

Tang (2009) studied the impact of monetary policy on economic growth in USA, UK and

Japan for 30 years. The study used Gross domestic product as the dependent variable and also

used money supply and its shock as independent variables coupled with the use of regression

technique, it was revealed that monetary policy and its shock has no significant effect on

economic growth. Hence, it was recommended that intermediate goals should be set by monetary

authorities and interest rate should be one of these goals.

Chih-Hsiang, Kam and Chan (2009) studied the effect of money supply on real output

and price in China between 1993 and 2008. The study used Gross domestic product and inflation

as dependent variables while inflation and money supply was used as independent variables

coupled with the use of Auto Regressive Integrated Moving Average (ARIMA) technique, it was

revealed that money supply has significant effect on economic growth. Hence, it was

recommended that tight monetary policy will help manage the overheated economy of China.

44
Table 2.1: Snapshot of Empirical Review from developed countries

Author(s) Objective Methodology Major Findings Country

Kutu, Nzimande Effect of ARDL Modeling Monetary policy China
and Msomi Monetary Policy technique has significant
(2017) on Growth of impact on
Industrial Sector economic growth
Borio and Effect of Generalized Monetary policy Developed
Gambacorta monetary policy Methods of has negative Countries
(2017) on bank lending Moments effect on bank
lending
Chan (2016) Effect of money Vector Error Money supply China
supply and Correction and inflation rate
inflation rate on Modeling have significant
economic growth Technique effect on
economic growth
Mushtaq and Effect of interest Panel ARDL Interest rate has Developed
Siddiqui (2016) rate on bank technique positive effect on countries
deposits bank deposit
Wong and Chong Effect of Ordinary Least Monetary policy Developed
(2015) Monetary Policy Square technique has significant Countries
on Economic effect on
growth economic growth
AbuDalu, Effect of ARDL technique Monetary policy Asia
Elsadig, exchange rate on variables have
Almasaled and economic growth positive effect on
Abuelgasim economic growth
(2014)
Douanla (2014) Effect of Generalized Monetary policy Franc Zone
monetary policy Methods of has both negative
on economic Moments and positive

45
 growth effect on
economic growth
Liu (2012) Effect of Descriptive and Monetary policy Australia
monetary policy Regression has significant
and bank size on technique effect on bank
bank lending lending
Ridhwan, Henri, Impact of OLS and VAR Monetary policy US and Europe
Nijkamp and Piet monetary policy technique has significant
(2010) on economic effect on
growth economic growth
Tang (2009) Impact of Regression Monetary policy USA, UK and
Monetary policy technique and its shock has Japan
on economic no significant
growth effect on
economic growth
Chih-Hsiang, Effect of money ARIMA Money supply China
Kam and Chan supply on real technique has significant
(2009) output and price effect on
economic growth
Source: Author’s Compilation (2017)

2.3.2 Empirical Review from Developing Countries

A number of studies have been conducted in developing countries of the world aimed at

investigating the impact of monetary policy on economic growth.

Drama (2017) studied the impact of monetary policy on economic growth in Cote

d‟Ivoire between 1990 and 2014. The study used Global oil price and world commodity price as

dependent variables and also used exchange rate, interest rate, money supply as independent

46
variables coupled with the use of Structural Vector Auto Regression technique, it was revealed

that monetary policy has insignificant effect on economic output. Hence, it was suggested that

the economy should be diversified and fiscal policy should be relied upon.

Srithilat and Sun (2017) studied the impact of monetary policy on economic development

in Lao PDR between 1989 and 2016. The study used Gross Domestic product as the dependent

variable and also used money supply, interest rate, inflation rate and exchange rate as

independent variables coupled with the use of Error Correction Modeling technique, it was

revealed that monetary policy has negative effect on economic growth. Hence, it was

recommended that monetary authorities should make use of exchange and interest rates as

monetary policy tools rather than money supply due to the inflationary effect.

Saqib and Aggarwal (2017) studied the effect of fiscal and monetary policy on economic

growth in Pakistan between 1984 and 2014. The study used Gross domestic product as the

dependent variable and also money supply and fiscal balance as independent variables coupled

with the use of Johansen Co-integration test, it was divulged that monetary and fiscal policy has

long run relationship with economic growth. Hence, it was recommended that monetary policy

should be given proper attention to foster economic growth.

Najaf (2017) studied the effect of monetary policy on economic growth of Pakistan

between 1982 and 2009. The study used Gross domestic product as the dependent variable and

also used liquidity ratio, money supply and cash ratio as independent variables coupled with the

use of the Vector Error Correction Modeling technique divulging that monetary policy has

significant effect on economic growth. Hence, it was divulged that operations of the financial

markets and employment should be improved upon to reduce inflation.

47
 Aftab, Mohsin and Saboor (2016) studied the impact of monetary policy on economic

growth in Pakistan between 1972 and 2011. The study used Gross domestic product as the

explained variable and also used money supply, inflation, exchange rate and interest rate as the

explanatory variables twinned with the use of Correlation and the Ordinary Least Square

technique divulging that monetary policy has significant effect on economic growth.

Bokreta and Benanaya (2016) studied the effect of fiscal and monetary policy on

economic growth in Algeria between 1970 and 2014. The study used Gross domestic Product per

capita as dependent variable and also used government expenditure, net taxes, exchange rate and

inflation rate as independent variables coupled with the use of the Vector Error Correction

Modeling technique; it was revealed that monetary policy has an insignificant effect while fiscal

policy has significant effect on economic growth. Hence, it was suggested that proper attention

should be given to monetary policy in stimulating economic growth and petrol prices should be

reduced.

Noman and Khudri (2015) studied the effects of monetary and fiscal policies on

economic growth in Bangladesh between 1979 and 2013. The study used gross domestic product

as the dependent variable and also used exchange rate, interest rate, inflation, narrow money,

government expenditure and government revenue as independent variables twinned with the

employment of the correlation and regression techniques, it was revealed that monetary and

fiscal policies affect economic growth. Hence, it was recommended that that proper attention

should be given to both policies to foster growth.

Asghar and Hussain (2014) studied the lags in the effect of monetary policy on inflation

in Pakistan between 1995 and 2008. The study used inflation as the dependent variable and used

48
growth rate of money supply and lagged values of money supply as independent variables

coupled with the use of the classical Ordinary Least Square technique, it was revealed that

money supply has a positive relationship with inflation while its lag after about six months has

positive relationship with inflation. Hence, it was recommended that the monetary policy of the

economy should be tightened to reduce inflation and spur economic growth.

Olweny and Chiluwe (2012) studied the effect of monetary policy on private sector

investment in Kenya between 1996 and 2009. The study used private sector investment as the

dependent variable and also used government gross domestic debt, money supply and treasury

bill rate as independent variables coupled with the use of the Vector Auto Regressive (VAR)

technique, it was revealed that monetary policy has a negative effect on private sector

investment. Hence, it was recommended that monitoring policy should narrow the spread

between lending and deposit rates.

Chaudhry, Qamber and Farooq (2012) studied the relationship between monetary policy

inflation and economic growth in Pakistan between 1972 and 2010. The study used Gross

Domestic Product as the explained variable and also used money supply, credit to private sector,

interest rate, consumer price index, exchange rate and budget deficit as independent variables

coupled with the use of descriptive statistics, correlation technique, Error Correction Modeling

and Granger Causality technique revealing that monetary policy has significant effect on

economic growth while there exists a bi-directional causality between exchange rate and

economic growth. Hence, it was recommended that the monetary authorities should maintain a

monetary tightening stance.

49
 Lashkary and Kashani (2011) studied the impact of monetary policy variables on

economic growth in Iran between 1959 and 2008. The study used Gross Domestic Product as the

dependent variable and also used Money volume, rate of employment, exchange rate,

government expenditure and government income as independent variables coupled with the use

of regression technique, it was divulged that monetary policy has significant effect on economic

growth. Hence, it was recommended that contractionary monetary policies should be used to

control inflation in Iran.

Irfan and Amen (2011) studied the impact of monetary policy on gross domestic product

in Pakistan between 1980 and 2009. The study used Gross Domestic Product as the dependent

variable and also used money supply, interest rate and inflation rate as independent variables

coupled with the use of regression technique; it was revealed that monetary policy has positive

effect on economic growth.

Mehdi and Reza (2011) studied the effect of monetary policy on industry sector growth in

Iran between 1961 and 2007. The study used real output as the dependent variable and also used

interest rate, exchange rate, credit to private sector, all share price index and consumer price

index as independent variables coupled with the use of the Auto Regressive Distributed Lag

modeling technique, it was revealed that monetary policy has significant effect on economic

growth. Hence, it was recommended that credit should be used prudently to promote investment

in the economy.

Poon (2010) tested the transmission mechanism of monetary policy in Malaysia between

1980 and 2004. The study used Gross domestic product as the dependent variable and also

employed exchange rate, share price, interest rate and bond rate as independent variables coupled

50
with the use of the Auto Regressive Distributed Lag modeling technique, it was revealed that

monetary policy has effect on economic growth using exchange rate and asset price as channels.

Table 2.2: Snapshot of Empirical Review from developing countries

Author(s) Objective Methodology Major Findings Country

Drama (2017) Impact of Structural VAR Monetary policy Cote d‟Ivoire
monetary policy technique has insignificant
on economic effect on
growth economic growth
Srithilat and Sun Impact of Error Correction Monetary policy Lao PDR
(2017) monetary policy Modeling has negative
on economic effect on
development economic growth
Saqib and Effect of fiscal Johansen Co- Monetary and Pakistan
Aggarwal (2017) and monetary Integration test fiscal policy has
policy on long run
economic growth relationship with
economic growth
Najaf (2017) Effect of Vector Error Monetary policy Pakistan
monetary policy Correction has significant
on economic Modeling effect on
growth technique economic growth
Aftab, Mohsin Impact of Correlation and Monetary policy Pakistan
and Saboor monetary policy Ordinary Least has significant
(2016) on economic Square technique effect on
growth economic growth
Bokreta and Effect of fiscal Vector Error Monetary policy Algeria
Benanaya (2016) and monetary Correction has insignificant
policy on Modeling effect on
economic growth technique economic growth

51
Noman and Effect of Correlation and Monetary and Bangladesh
Khudri (2015) monetary and regression fiscal policy has
fiscal policies on technique significant effect
economic growth on economic
growth
Asghar and Lags in the effect Ordinary Least Money supply Pakistan
Hussain (2014) of monetary Square technique has positive
policy on relationship with
inflation inflation while its
lag after six
months has
positive
relationship with
inflation
Olweny and Effect of Vector Auto Monetary policy Kenya
Chiluwe (2012) monetary policy Regressive has negative
on private sector (VAR) technique effect on private
investment sector investment
Chaudhry, Relationship Descriptive Monetary policy Pakistan
Qambar and between Statistics, has significant
Farooq (2012) monetary policy, Correlation effect on
inflation and analysis and economic growth
economic growth Error Correction
Modeling
Lashkary and Impact of Regression Monetary policy Iran
Kashani (2011) monetary policy technique has significant
variables on effect on
economic growth economic growth
Irfan and Amen Impact of Regression Monetary policy Pakistan
(2011) monetary policy technique has positive
on economic effect on

52
 growth economic growth
Mehdi and Reza Effect of ARDL technique Monetary policy Iran
(2011) monetary policy has significant
on industry effect on
sector growth economic growth
Poon (2010) Transmission ARDL technique Monetary policy Malaysia
mechanism of has significant
monetary policy effect on
economic growth
Source: Author’s Compilation (2017)

2.3.3 Empirical Review from Nigeria

Brilliant scholars in Nigeria have in various attempts in literature considered the impact

of monetary policy on economic growth.

Inam and Ime (2017) studied the impact of monetary policy on economic growth in

Nigeria between 1970 and 2012. The study used Gross Domestic Product as the dependent

variable and also used money supply, inflation rate, government expenditure, interest rate,

exchange rate, population and investment rate as independent variables coupled with the use of

the Ordinary Least Square regression technique, it was divulged that monetary policy has

insignificant effect on economic growth. Hence, it was recommended that government should

increase productive population to assist in achieving economic growth.

Anowor and Okorie (2016) studied the impact of monetary policy on economic growth in

Nigeria between 1982 and 2013. The study used Gross Domestic Product as the dependent

variable and also used Interest rate; cash reserve ratio and monetary policy ratio as independent

variables coupled with the use of the classical ordinary least square and the error correction

53
modeling technique revealing that monetary policy has a positive effect on economic growth.

The study recommended that attention should be given to cash reserve ratio in a bid to stabilize

the economy.

Akinjare-Victoria, Babajide and Okafor (2016) studied the effect of monetary policy on

economic growth in Nigeria between 1970 and 2013. The study used Gross Domestic Product as

the dependent variable and also used exchange rate, interest rate, money supply and inflation rate

as independent variables coupled with the use of Error Correction Mechanism, it was divulged

that monetary policy has significant effect on economic growth. Hence, it was recommended that

monetary policies should be used to create a favourable climate for investment and economic

growth.

Obadeyi, Okhiria and Afolabi (2016) studied the impact of monetary policy on economic

growth in Nigeria between 1990 and 2012. The study used gross domestic product as the

dependent variable and also used inflation, interest rate, money supply and exchange rate as the

independent variable coupled with the use of the classical Ordinary Least Square and Error

Correction technique, it was revealed that monetary policy has negative effect on economic

growth. Hence, it was recommended that monetary authorities should make use of

countercyclical policy that will lead to the expansion of the economy.

Osmond, Egbulonu and Emerenini (2015) studied the impact of monetary policy

variables on the manufacturing sector of Nigeria between 1981 and 2012. The study used

Industry contribution to Gross Domestic Product as the dependent variable and also used Interest

rate, credit to private sector, money supply and inflation as independent variables coupled with

the use of Error Correction Modeling, it was revealed that monetary policy has significant effect

54
on manufacturing sector of economy. Hence, it was revealed that monetary authorities should

avoid inconsistencies in the implementation of policies.

Baghebo and Ebibai (2014) studied the impact of monetary policy on economic growth in

Nigeria between 1980 and 2011. The study used Gross Domestic Product as the dependent

variable and also used money supply, liquidity ratio and cash ratio as the independent variable

coupled with the use of the Error Correction Modeling technique, it was revealed that monetary

policy has positive effect on economic growth. Hence, it was recommended that the monetary

authorities should introduce more flexible monetary instruments that can meet the ever growing

financial sector.

ThankGod and Karimo (2014) studied the effect of monetary policy on economic growth

and inflation in Nigeria between 1970 and 2011. The study used Gross Domestic Product as the

dependent variable and also used Interest rate, consumer price index and money supply as

independent variable coupled with the use of VAR technique revealing that monetary policy

has a long run effect on output growth. Hence, it was recommended that policy makers should

control inflation and adopt expansionary polices to maintain growth.

Sulaiman and Migiro (2014) studied the nexus between monetary policy and economic

growth in Nigeria between 1981 and 2012. The study used Gross Domestic Product as the

dependent variable and also used Cash reserve ratio, money supply, interest rate and monetary

policy rate as the independent variables coupled with the use of the Johansen Co-Integration test

and the Engle Granger Causality test, it was revealed that there exists a long run relationship

between monetary policy and economic growth and that monetary policy causes economic

55
growth. Hence, it was recommended that the supervisory framework should be strengthened to

assist the effectiveness of the monetary policy framework.

Amassoma, Nwosa and Olaiya (2011) studied the effect of monetary policy on

macroeconomic stabilization in Nigeria between 1986 and 2009. The study used interest rate as

the dependent variable and also used broad money supply, inflation, exchange rate and gross

domestic product as independent variables coupled with the use of Ordinary Least Square and

Johansen Co-integration test, it was discovered that monetary policy has a long run but

insignificant relationship with macroeconomic stabilization. Hence, it was recommended that

monetary authorities should introduce greater flexibility in the monetary policy implementation.

Okwu, Obiakor, Falaiye and Owolabi (2011) studied the effect of monetary policy

innovations on stabilization of commodity prices in Nigeria between 1995 and 2009. The study

used commodity prices as the dependent variable and also used monetary policy rate and the

broad money aggregate as independent variables coupled with the use of Ordinary Least Square

technique, it was revealed that monetary policy has a positive relationship with commodity

prices. Hence, it was suggested that monetary authorities should fine-tune their policies regularly

to ensure relevance and consistency.

Amassoma, Nwosa and Ofere (2011) studied the nexus between interest rate

deregulation, lending rate and agricultural productivity in Nigeria between 1986 and 2009. The

study used Agricultural output as the dependent variable and also used bank lending, credit to

agricultural sector, credit to private sector, direct investment, exchange rate and interest rate as

independent variables coupled with the use of Error Correction Modeling, it was revealed that

interest rate deregulation has no effect on agricultural productivity in Nigeria. Hence, it was

56
suggested that agricultural insurance scheme should be developed to boost productivity in

Nigeria.

Ogunmuyiwa and Ekone (2010) studied the relationship between money supply as a

monetary policy variable on economic growth in Nigeria between 1980 and 2006. The study

used Gross domestic Product as the dependent variable and also used Money Supply as the

independent variable coupled with the use of the Ordinary Least Square and the Error Correction

Mechanism; it was revealed that money supply is positively related to economic growth. The

study recommended that monetary authorities should harmonize the contractionary and

expansionary policies to reduce the rate differential between the productive and unproductive

credit supplied in the economy.

Table 2.3: Snapshot of Empirical Review from Nigeria

Author(s) Objective Methodology Major Findings Country

Inam and Ime Impact of Ordinary Least Monetary policy Nigeria
(2017) monetary policy Square has insignificant
on economic Technique effect on
growth economic growth
Anowor and Impact of Ordinary Least Monetary Policy Nigeria
Okorie (2016) monetary policy Square technique has positive
on economic and Error effect on
growth Correction economic growth
Modeling
Akinjare- Effect of Error Correction Monetary policy Nigeria
Victoria, monetary policy Modeling has significant
Babajide and on economic effect on
Okafor (2016) growth economic growth
Obadeyi, Okhiria Impact of Ordinary Least Monetary policy Nigeria

57
and Afolabi monetary policy Square and Error has negative
(2016) on economic Correction effect on
growth Modeling economic growth
Osmond, Impact of Error Correction Monetary policy Nigeria
Egbulonu and monetary policy Modeling has significant
Emerenini variables on the effect on
(2015) manufacturing manufacturing
sector sector of the
economy
Baghebo and Impact of Error Correction Monetary policy Nigeria
Ebibai (2014) monetary policy Modeling has positive
on economic effect on
growth economic growth
ThankGod and Effect of Vector Auto Monetary policy Nigeria
Karimo (2014) monetary policy Regressive has long run
on economic technique effect on output
growth and growth
inflation
Sulaiman and The nexus Johansen Co- Monetary policy Nigeria
Migiro (2014) between Integration test has a long run
monetary policy and Engle relationship with
and economic Granger economic growth
growth Causality test while monetary
policy causes
economic growth
Amassona, Effect of Ordinary Least Monetary policy Nigeria
Nwosa and monetary policy Square and has a long run
Olaiya (2011) on Johansen Co- but insignificant
macroeconomic Integration test relationship with
stabilization macroeconomic
stabilization

58
Okwu, Obiakor, Effect of Classical Monetary policy Nigeria
Falaiye and monetary policy Ordinary Least has a positive
Owolabi (2011) innovations on Square relationship with
stabilization of commodity
commodity prices
prices
Amassoma, Nexus between Error Correction Interest rate Nigeria
Nwosa and Ofere interest rate Modeling deregulation has
(2011) deregulation, no effect on
lending rate and agricultural
agricultural productivity
productivity
Ogunmuyiwa Relationship Ordinary Least Money supply is Nigeria
and Ekone between money Square and Error positively related
(2010) supply as a Correction to economic
monetary policy Mechanism growth
variable on
economic growth
Source: Author’s Compilation (2017)

2.4 Gaps Identified

A proper cross examination of contributions in literature revealed some gaps which this

study intends to fill. First of all, it was discovered that there exists mixed result in literature as

regards the subject matter as some studies revealed a positive impact while some others revealed

a negative impact; as a result, this study is set to consider the subject matter to uncover the real

fact. Moreso, most studies failed to establish the direction of causality between the monetary

policy indices and economic growth. Hence, this study is set to adopt the Engle Granger

causality approach to examine the causal relationship among the variables in the study.

59
 CHAPTER THREE

RESEARCH METHOD

3.1 Theoretical Framework

From the foregoing, it is imperative to note that this current study theoretically underpin

the working of a theory among those specified above in the investigation of the impact of

monetary policy on economic growth. Intuitively, the study considers one of the theories relating

to this subject matter as the best suitable for the theoretical framework due to its close affiliation

with the broad objective of the study. The study relied on the Endogenous Growth theory as the

underpinning theory underlying the research work. The proponents of this theory assume that

factors leading to the growth of an economy are endogenous in nature which points to the fact

that the actions and inactions of monetary authorities through their various monetary policies

within the country influences economic growth. Hence, monetary policy must be considered as a

very serious matter that is essential to the growth of every economy.

3.2 Model Specification

In a bid to capture the impact of monetary policy on economic growth in these countries,

this current study employed an empirical model which is built based on the slight modification of

the model used in the study carried out by Obadeyi, Okhiria and Afolabi (2016). The model

which can be expressed mathematically in its original form as:

GDP = f (INFR, INTR, MS, EXGR)

Where:

GDP = Gross Domestic Product

INFL = Inflation Rate

60
INTR = Interest Rate

MS = Money Supply

EXGR = Exchange Rate

However, after modifications by the researcher in a bid to bridge more knowledge gaps

by considering the impact of another monetary policy variable on economic growth, the modified

model is stated below as:

GDP = f (MPR, INFL, INTR, MS, EXGR, µ)…………………………………………. (3.1)

This model can for the purpose of simplicity be stated in the econometric form of

equation as depicted below:

GDP = B0 + B1MPR + B2INFL + B3INTR + B4MS + B5EXGR + µ…………………… (3.2)

Where;

GDP = Gross Domestic Product

MPR = Monetary Policy Rate

INFL = Inflation Rate

INTR = Interest Rate

MS = Money Supply

EXGR = Exchange Rate

µ = Error Term

61
B0 = Constant Parameter

B1 – B5 = Coefficients of Regression

In a bid to avoid spuriousity in estimation, the study moves an inch by process of loglinearization

and the entire and the entire model becomes:

Log (GDP) = B0 + B1Log(MPR) + B2Log(INFL) + B3Log(INTR) + B4Log(MS) +

B5Log(EXGR) + µ………………………………………… (3.3)

Where:-

Log = Natural Logarithm

From equation 3 above, the research model can further be stated in time series form as depicted

below:-

Log(GDP)t = B0+ B1Log(MPR)t + B2Log(INFL)t + B3Log(INTR)t + B4Log(MS)t +

B5Log(EXGR)t + µ…………………………………………… (3.4)

Where:-

t = Time Lag

Again, by formulating the Error Correction Model (ECM) for the Auto Regressive

Distributed Lag (ARDL) Model as obtained from equation (3), the model becomes:

62
 ( )

∑ ( ) ∑ ( )

∑ ( ) ∑ ( ) ∑ ( )

∑( ) ∑ ( )

Where: -

Δ - Change

t-1 - Lagged value of each variables

∑t - White noise/residual

∑ ( ) - Error Correction term

In testing for the existence of long run equilibrium relationship, the error correction

model i.e. equation (3.5) can be conducted by placing some restrictions on estimated long run

coefficient of variables. Therefore, the hypothesis for the test is formulated as follows:

H0: B1 = B2 = B3 = B4 = B5 = 0 (There is no long run equilibrium relationship or No co-

integration)

H1: B1 ≠ B2 ≠ B3 ≠ B4 ≠ B5 ≠ 0 (There is the presence of long run equilibrium relationship or

co-integration exist)

3.3 Re-Statement of Hypothesis

The hypothesis to be tested in this study as stated earlier in the introductory part of the

study can be restated as:

H0: Monetary policy has no significant impact on economic growth in selected countries.

H1: Monetary Policy has significant impact on economic growth in selected countries.

63
3.4 Estimation Techniques

This current study in its bid to examine the short and long-run dynamic relationship

between the variables in the model employs the Johansen‟s co-integration framework with

respect to the ECM model specified earlier in this chapter of the study provided that all variables

were found to be integrated in the same order either at level, first difference or second difference.

If co-integration exists alongside its extents and forms, the next step required is to develop an

over-parameterized Error Correction Model (ECM 1) which involves the process of leading and

lagging the variables and then a parsimonious error correction model (ECM 2) that incorporates

long-run equilibrium relationship and short-run dynamics into the model will be developed.

However, if the variables were found to be integrated in different order as revealed by the Unit

Root Test, the study was not be able to use the Co-Integration Test which can only be used when

variables are integrated in the same order necessitating the need for the use of the Auto

Regressive Distributed Lag (ARDL) model which can also be premised on the ECM model

developed in the study (Oteng-Adayie & Frimpong, 2006).

The estimation technique will however follow a three-step procedure:

i. The time series properties will be investigated and the order of integration is

determined using the Augmented Dickey-Fuller (ADF) unit root test. The presence of

unit roots culminated in the need to further test for co-integration between variables.

ii. After the unit root test, if the variables are found to be integrated of same order, we

apply the Juselieus and Johansen co-integration test to determine the number of co-

integrating vectors. In this case, trace and maximum Eigen-value test is applied.

Trace test defines whether the number of co-integrating vector is zero or one and then

Maximum Eigen value test is used to define whether a single co-integration equation

64
 is sufficient or not. However, in case of mixed integration, the Auto Regressive

Distributed Lag (ARDL) model will be adopted.

On the other hand, if the variables are found to non-stationary, we make them stationary

through differencing before determining the number of co-integrating vectors. For example, if

some variables are not stationary at level, we can differentiate the variables in order to make

them stationary.

iii. If the variables are found to be co-integrated, the next step is to develop an Over-

parameterized ECM i.e. ECM1 and a parsimonious ECM (ECM2) that incorporates

long run equilibrium relationship and the short run dynamics.

The choice of the technique rests solely on the fact to establish a long run equilibrium

relationship between the dependent (explained) variable and the independent (explanatory)

variables unlike the Ordinary Least Square (OLS) analysis which is susceptible to spuriousity of

results and of which its results are short run oriented and somewhat misleading.

3.5 Econometric Tests

3.5.1 Generalized Method of Moments (GMM)

In a bid to examine the effect of monetary policy lag on economic growth, the dynamic

panel data analysis of which the GMM is included is employed. The use of the methodlogy

allows for the provision of more information and precision in the analysis. According to Arellano

and Bond (1991), in this analysis, the lag of the dependent variable is also included as an

explanatory variable. Hence, the model specification for the GMM model is stated as:

GDPit = B0 + B1GDPit-1 + B2MPRit + B3MPRit-1 + B4INFLit + B5INFLit-1 + B6INTRit + B7INTRit-1

+ B8MSit + B9MSit-1 + B10EXGRit + B11EXGRit-1 + µ……………………… (3.6)

65
Where: -

It - Contemporaneous or current value of each variable

it-1 - Lagged value of each variables

µ - White noise/residual

3.5.2 Unit Root Test (URT)

Following the submission made by Engle and Granger (1985) and Dickey and Fuller

(1981), there is the likelihood of obtaining a spurious regression if the series that generate the

results are non-stationary. The Unit root test is a standard approach in co-integration analysis

used for determining and re-determining the stationarity of the time series properties of the data.

This is performed using the Augmented Dickey Fuller (ADF) unit root test.

3.5.3 Co-Integration Test

Co-integration implies that if two or more series are linked to form an equilibrium

relationship spanning the long run, even though the series themselves may be non-stationary,

they will move closely together over time and their difference will be stationary. The Johansen-

Juselius (JJ) maximum likelihood method of co-integration test is adopted in this study simply to

show the long-run relationship subsisting between the dependent and the independent variables.

This is done by evaluating both the trace and maximum Eigen statistics to determine the co-

integration rank. The test is conducted assuming a linear deterministic trend with a lag interval of

1 to 4.

3.5.4 Standard Error Test (S.E Test)

The standard error test is done to determine the significance of each independent variable

in the explanation of the behaviour of the dependent variable. It is done using the standard error

statistics obtained from the co-integration equation of the co-integration test. However, the test is

66
a measure of dispersion of the estimate around the true parameter judges the statistical reliability

of the parameters in the long run. An alternative method to test for the statistical significance of

parameters is the Probability Value Test, where the significance of a parameter can be obtained

only if the p-value is less than or equal to 0.05 (5%) i.e. p-value ≤ 0.05, here the variable is

statistically significant and if otherwise, the variable is statistically insignificant.

3.5.5 Coefficient of Multiple Determination (R2)

The coefficient of multiple determination (R2) is used to measure the degree at which the

behaviour of the dependent variable is explained by the independent variables. It also takes into

account the measurement of the behaviour that is not explained by the model (white noise), i.e.

the extent to which the regression fails to explain the dependent variable (measure of

noise/disturbance). More tensely, it measures the extent to which the total variation of the

endogenous variable is explained by the exogenous variables.

3.5.6 Overall Significance of the Model (F-Test)

The F-test is employed to reveal the overall significance of the entire model to show if

the model adopted and modified is statistically significant or not. This is done on a tail test with

the comparison of the table value to the estimated value of F statistics. It is often done using a

two-tailed test.

3.5.7 Durbin Watson Test (DW-Test)

The Durbin-Watson test is used to determine whether or not there is presence of

Autocorrelation or serial correlation in a model adopted. The D*W statistics could either fall in

the area of acceptance (no autocorrelation) region, positive or negative autocorrelation region or

the inconclusive region depending on the area which the D*W statistical value falls within the

67
Durbin Watson graph. However, in the presence of auto correlation, auto correlation will be

corrected using the Cochrane Ocrutt Iterative method.

3.5.8 Auto Regressive Distributed Lag (ARDL) Model

The Auto Regressive Distributed Lag (ARDL) bounds testing methodology as developed

by Pesaran and Shin (1999) has been preferred to the co-integration analysis developed by Engle

and Granger (1987) and Johansen and Juselius (1990) due to the low power problems associated

with the co-integration analysis. Furthermore, the ARDL can be used to test for the significance

of the lagged levels of the variables considered (Pesaran, Shin & Smith,2001) and it is used

irrespective of the order of integration either in a case of mixed integration of pure integration

circumventing problems resulting from non-stationarity of data (Shrestha & Chowdhury, 2007).

However, the ARDL model can be established premised on the ECM model earlier stated in the

study as the ECM model adopted prevents spurious results because it explains the previous

disequilibrium in current period. Obviously, the ARDL technique is preferred to the conventional

co-integration technique due to some reasons. According to Giles (2013), it is the most recent

estimation technique to estimate long run and short run dynamics, it allows different variables to

be assigned different lags in the model, it can accommodate more than two lags and up to six

variables, it involves just a single equation set up, making it easy to implement and interpret.

Also, Katircioglu (2009) assumes that it can be used with a mixture of integrated variables at

different orders, that is, I(0) and I(1) while Dritsakis (2011) posited that it allows for short run

and long run model to be estimated simultaneously. Meanwhile, Narayan (2005) opined that it is

good for both small and large sample size. The ARDL analysis was carried out in the study using

the E-Views 9 statistical package.

68
3.5.9 Model Stability and Diagnostic Test

In a bid to check the validity of the model, several tests as recommended by Pesaran,

Shin and Smith (2001) will be conducted, such tests include Serial correlation test (Breusch -

Godfrey LM Test) used to test for the correlation or correspondence of residuals in the model,

the Normality Test (Jarque-Bera Test) used to test for the normality of distribution of the model

residuals and the Heteroskedasticity test to test for the presence of variance of errors across

observations. The tests will be carried out using the E-Views 9 Statistical package.

3.5.10 Engle Granger Causality Test

In a bid to examine the short run direction of causal relations (causality) between

monetary policy and economic growth, the granger causality test was employed with an optimal

lag of 2. The temporal granger causality test according to Granger (1969) is a statistical

hypothesis test that is used to bridge the limitation of the explanation of the causes of one

variable by another. It shows whether the past (lagged) values of the identified explanatory

variables truly cause the behaviour of the explained variable. It helps to know whether a past

change in one variable, say X causes a current change in another variable Y or whether the

relation works in the opposite direction. This can either be Unidirectional, Bidirectional, or no

Relation at all as revealed in related studies of several authors. It is tested using the Probability

value and the F-statistics. However, the Granger equations for the research model are presented

as;

∑ ∑ ( )

69
 ∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

∑ ∑ ( )

3.6 Sources of Data

The study was conducted using a time series annual data for the selected countries viz;

Nigeria, USA and UK spanning through 1986 to 2016. The choice of this period is premised

upon the fact that it strictly marks the beginning of the reform era relating to liberalization in the

70
Nigeria. The data needed in carrying out this research work which is secondary in nature were

sourced from the database of the World Bank.

These data were painstakingly extracted from the soft copy of the above named sources

as made available by the relevant authorities and via the internet. They were carefully analysed

and where necessary, some appropriate calculations were made based on the derivation formula

relating to the variables given in literature.

3.7 Description of Variables

3.7.1 Gross Domestic Product

The Gross Domestic Product shows the level of the country‟s output. It is an indicator of

the economic performance of a country as it was used by Chan (2016) as a proxy for economic

growth which represents the dependent variable for the study.

3.7.2 Monetary Policy Rate

This is the rate at which credit is extended to financial institutions. It is used in the

variable because it is a method of indirect control by monetary authorities in controlling the

lending behaviour of banks and the rate of credit supplied to the economy at large. It was used by

Sulaiman and Migiro (2014) as an independent variable.

3.7.3 Inflation Rate

Inflation is the persistent rise in the general price level in an economy in a given period.

In the model, inflation is used as a proxy for macroeconomic instability that may arise within the

economy as a result of the actions and inactions of monetary authorities. Amassoma, Nwosa and

Olaiya (2011) used inflation rate as an explanatory variable in his study.

71
3.7.4 Interest Rate

Interest rate is the opportunity cost of holding money, if it is set at a high rate, the cost of

borrowing becomes high and it then affects economic growth. It was used by Amassoma, Nwosa

and Ofere (2011) as an independent variable in their study.

3.7.5 Money Supply

This is the total credit made available or supplied to the economy as a whole at a

particular time for various purposes. It was used by Ogunmuyiwa and Ekone (2010) as an

independent variable in their study.

3.7.6 Exchange Rate

Exchange rate is the rate at which domestic currency can be converted to a foreign

currency. It affects the rate of import and exports as a well as the economy as a whole. It was

used as an independent variable by AbuDalu, Elsadig, Almasaled and Abuelgasim (2014).

3.8 Theoretical Expectation

Theoretical expectations are determined by the underlying principles of intuitive

economic theory and they refer to the expected relationship between the controlled variable and

the control variables.

In this study, it is expected that monetary policy rate can have both positive and negative

effects on economic growth based on the economy. This can be denoted mathematically as: f‟

(MPR)>< O, which means that a unit increase in monetary policy rate will increase or decrease

Gross Domestic Product by the same unit.

Also, there can be a positive or negative relationship between inflation and Gross

Domestic Product in that an increase in inflation can increase or decrease economic growth. This

direct relationship can be stated as: f‟ (INFL)><0. Then, interest rate can have a negative or

72
positive relationship with economic growth, hence, an increase in interest rate will lower or spur

economic growth, this relationship can be expressed as: f‟ (INTR)><0. Furthermore, money

supply can be positively or negatively related with economic growth as an increase in money

supply can improve or lower economic growth, this can be stated as f‟ (MS)><0. Moreso,

exchnage rate is expected to be positively or negatively related to economic growth in that an

increase in exchange rate can spur or reduce economic growth and this can be expressed as f‟

(EXGR)><0.

3.9 Limitation of Data

The data that will be analysed in the current study is the data for the economy of the

selected countries viz; Nigeria, USA and UK as obtained from secondary sources. The scope for

the study spanned from 1986 to 2016, as the time series annual data relating to all the variables

for the period will be used.

The major limitation in this regard is the aspect of dearth of literature on the subject

matter especially recent contributions from developed countries of the world. Also, financial

constraint was also another major factor of limitation. Nonetheless, irrespective of the

limitations, the study was conducted to produce significantly reliable and objective results that

can be useful for further research as well as assisting the monetary authorities in their policy

formulation.

73
 CHAPTER FOUR

DATA ANALYSIS AND FINDINGS

This study examined the impact of monetary policy and its lag on economic growth in

Nigeria, UK and USA between 1986 and 2016 with the use of Auto Regressive Distributed Lag

(ARDL) model to assess the long run impact in the presence of mixed integration order. In line

with the research model used for the study, Gross Domestic Product (GDP) was used as a proxy

for economic growth which is the dependent variable while Monetary Policy Rate (MPR),

Inflation (INFL), Interest Rate (INTR), Money Supply (MS) and Exchange Rate (EXGR) were

used as the explanatory variables. However, the study first examined the Generalized Method of

Moments (GMM) technique to look into the effect of monetary policy lag for the three countries

(Nigeria, USA and UK) before proceeding to the Augmented Dickey Fuller Unit Root Test, the

Auto Regressive Distributed Lag modeling technique and the Engle Granger Causality test. In

the light of this, this part of the study is exclusively reserved for the analysis and interpretation of

the research findings.

4.1 Data Presentation

The raw and log-linearized data used in the study were secondary data spanning from

1986 to 2016 culled and analysed and is duly presented as shown in table A (1& 2) in the

appendix.

4.2 Generalized Method of Moments (GMM)

The result of the GMM technique to examine the effect of monetary policy lag on

economic growth as presented in table B of Appendix II, III and IV is summarized below:

74
4.2.1 Nigeria

Table 4.1 GMM Result for Nigeria

Dependent Variable: GDP

Variable Coefficient Std. Error t-Statistics Prob.

GDP(-1) 0.844037 0.087529 9.642913 0.0000

MPR -0.011534 0.023149 -0.498225 0.6244
MPR(-1) -0.050199 0.037712 -1.331101 0.1998
INFL -0.042981 0.012816 -3.353727 0.0035
INFL(-1) 0.016762 0.011385 1.472237 0.1582
INTR 0.032350 0.022029 1.468488 0.1592
INTR(-1) -0.033883 0.040483 -0.836984 0.4136
MS -0.118146 0.074160 -1.593129 0.1285
MS(-1) 0.147576 0.085949 1.717024 0.1031
EXGR 0.016802 0.022311 0.753093 0.4611
EXGR(-1) -0.113942 0.044501 -2.560424 0.0197
C 4.001395 2.194582 1.823306 0.0849
2
R =0.991638 DW-Stat. = 1.911456
Source: Author‟s Computation (2017) using E-Views 9 (See Table B in Appendix II)
From the table 4.1 above, in Nigeria, the constant parameter is positively signed to the

tune of 4.001395 units which implies that if all variables are held constant (i.e. at zero level),

Gross Domestic Product (GDP) will increase by 4.001395 units. Also, the lagged value of Gross

Domestic Product (GDP(-1)), lagged value of Inflation (INFL(-1)), Interest Rate (INTR), lagged

value of Money Supply (MS(-1)) and Exchange Rate (EXGR) all have positive effect on

economic growth to the tune of 0.844037, 0.016762, 0.032350, 0.147576 and 0.016802 units

respectively which implies that a unit increase in these variables will increase economic growth

in Nigeria by same units.

On the other hand, Monetary Policy Rate (MPR), the lagged value of Monetary Policy

Rate (MPR(-1)), Inflation (INFL), lagged value of Interest Rate (INTR(-1)), Money Supply (MS)

75
and lagged value of Exchange Rate (EXGR(-1)) all have negative effect on economic growth to

the tune of -0.011534, -0.050199, -0.042981, -0.033883, -0.118146 and -0.113942 units

respectively which implies that a unit increase in the variables specified above will reduce the

rate of economic growth by same units in Nigeria.

4.2.2 United States of America (USA)

Table 4.2 GMM Result for USA

Dependent Variable: GDP

Variable Coefficient Std. Error t-Statistics Prob.

GDP(-1) 1.001684 0.033121 30.24354 0.0000

MPR -0.004747 0.033745 -0.140683 0.8899
MPR(-1) -0.044892 0.033052 -1.358232 0.1932
INFL -0.002150 0.004629 -0.464507 0.6485
INFL(-1) 0.000815 0.004056 0.200945 0.8433
INTR 0.011918 0.021319 0.559022 0.5839
INTR(-1) 0.001364 0.017505 0.077892 0.9389
MS 0.229194 0.074215 3.088262 0.0071
MS(-1) -0.265897 0.083217 -3.195218 0.0056
EXGR 0.044203 0.067413 0.655707 0.5213
EXGR(-1) -0.022195 0.041615 -0.533333 0.6011
C 1.020885 0.438411 2.328600 0.0333
2
R =0.998986 DW-Stat. = 2.842250
Source: Author‟s Computation (2017) using E-Views 9 (See Table B in Appendix III)
From the table 4.2 above, in USA, the constant parameter is positively signed to the tune

of 1.001684 units which implies that if all variables are held constant (i.e. at zero level), Gross

Domestic Product (GDP) will increase by 1.001684 units. Also, the lagged value of Gross

Domestic Product (GDP(-1)), lagged value of Inflation (INFL(-1)), Interest Rate (INTR), lagged

value of Interest Rate (INTR(-1)), Money Supply and Exchange Rate (EXGR) all have positive

effect on economic growth to the tune of 1.001684,0.000815, 0.011918, 0.001364, 0.229194 and

76
0.044203 units respectively which implies that a unit increase in these variables will increase

economic growth in USA by same units.

On the other hand, Monetary Policy Rate (MPR), the lagged value of Monetary Policy

Rate (MPR(-1)), Inflation (INFL), lagged value of Money Supply (MS(-1)) and lagged value of

Exchange Rate (EXGR(-1)) all have negative effect on economic growth to the tune of -

0.004747, -0.044892, -0.002150, -0.265897 and -0.022195 units respectively which implies that

a unit increase in the variables specified above will reduce the rate of economic growth by same

units in USA.

4.2.3 United Kingdom (UK)

Table 4.3 GMM Result for UK

Dependent Variable: GDP

Variable Coefficient Std. Error t-Statistics Prob.

GDP(-1) 1.122529 0.338671 3.314513 0.0211

MPR -0.013038 0.043772 -0.297868 0.7778
MPR(-1) -0.038857 0.030101 -1.290861 0.2532
INFL -0.002953 0.010954 -0.269628 0.7982
INFL(-1) 0.003916 0.007993 0.489972 0.6449
INTR 0.012237 0.007611 1.607793 0.1688
INTR(-1) 0.002220 0.009531 0.232951 0.8250
MS 0.013668 0.052918 0.258290 0.8065
MS(-1) -0.086118 0.031568 -2.728050 0.0414
EXGR 0.086455 0.089642 0.964453 0.3791
EXGR(-1) 0.083411 0.082760 1.007860 0.3598
C -2.178914 7.551061 -0.288557 0.7845
R2=0.999179 DW-Stat. = 2.906495
Source: Author‟s Computation (2017) using E-Views 9 (See Table B in Appendix IV)

77
 From the table 4.2 above, in UK, the constant parameter is negatively signed to the tune

of -2.178914 units which implies that if all variables are held constant (i.e. at zero level), Gross

Domestic Product (GDP) will reduce by 2.178914 units. Also, the lagged value of Gross

Domestic Product (GDP(-1)), lagged value of Inflation (INFL(-1)), Interest Rate (INTR), lagged

value of Interest Rate (INTR(-1)), Money Supply, Exchange Rate (EXGR) and lagged value of

Exchange Rate (EXGR(-1)) all have positive effect on economic growth to the tune of 1.122529,

0.003916, 0.012237, 0.002220, 0.013668, 0.086455 and 0.083411 units respectively which

implies that a unit increase in these variables will increase economic growth in the UK by same

units.

On the other hand, Monetary Policy Rate (MPR), the lagged value of Monetary Policy

Rate (MPR(-1)), Inflation (INFL) and lagged value of Money Supply (MS(-1)) all have negative

effect on economic growth to the tune of -0.013038, 0.038857, -0.002953 and -0.086118 units

respectively which implies that a unit increase in the variables specified above will reduce the

rate of economic growth by same units in UK.

Summarily, it can be deduced that monetary policy rate, its lagged value and inflation has

negative effect on economic growth in the three countries. However, Money supply and its

lagged value had a negative and positive effect on economic growth in Nigeria while money

supply lag had negative effect on economic growth in USA and UK while money supply has

positive effect on economic growth in both countries.

78
4.3 Test for the Significance of the Parameters in the Short Run (Probability Test)

In a bid to test for the statistical significance of the variables in the study, the probability

test will be employed for this purpose. This is done by considering the probability value attached

to the variables as presented in the GMM results.

Decision Rule:

If the probability value attached to the variable is less than 0.05 i.e. 5% significant is

statistically significant but if otherwise, the variable is considered to be statistically insignificant.

4.3.1 Nigeria

Table 4.4: Short Run Probability Test (Nigeria)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Significant
GDP(-1) 0.844037 0.0000
Insignificant
MPR -0.011534 0.6244
Insignificant
MPR(-1) -0.050199 0.1998
Significant
INFL -0.042981 0.0035
Insignificant
INFL(-1) 0.016762 0.1582
Insignificant
INTR 0.032350 0.1592
Insignificant
INTR(-1) -0.033883 0.4136
Insignificant
MS -0.118146 0.1285

79
 Insignificant
MS(-1) 0.147576 0.1031
Insignificant
EXGR 0.016802 0.4611
Significant
EXGR(-1) -0.113942 0.0197
Source: Author‟s Computation (2017) (See GMM result computed in table B of Appendix II)

From the table 4.4 above, it can be deduced that the lagged value of Gross Domestic

Product (GDP(-1)), Inflation and the lagged value of Exchange rate (EXGR(-1)) exert significant

effect on economic growth in Nigeria.

4.3.2 United States of America

Table 4.5: Short Run Probability Test (USA)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Significant
GDP(-1) 1.001684 0.0000
Insignificant
MPR -0.004747 0.8899
Insignificant
MPR(-1) -0.044892 0.1932
Insignificant
INFL -0.002150 0.6485
Insignificant
INFL(-1) 0.000815 0.8433
Insignificant
INTR 0.011918 0.5839
Insignificant
INTR(-1) 0.001364 0.9389

80
 Significant
MS 0.229194 0.0071
Significant
MS(-1) -0.265897 0.0056
Insignificant
EXGR 0.044203 0.5213
Insignificant
EXGR(-1) -0.022195 0.6011
Source: Author‟s Computation (2017) (See GMM result computed in table B of Appendix III)

From the table 4.5 above, it can be deduced that the lagged value of Gross Domestic

Product (GDP(-1)), Money Supply (MS) and the lagged value of Money Supply (MS(-1)) exert

significant effect on economic growth in the USA.

4.3.3 United Kingdom

Table 4.6: Short Run Probability Test (UK)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Significant
GDP(-1) 1.122529 0.0211
Insignificant
MPR -0.013038 0.7778
Insignificant
MPR(-1) -0.038857 0.2532
Insignificant
INFL -0.002953 0.7982
Insignificant
INFL(-1) 0.003916 0.6449
Insignificant
INTR 0.012237 0.1688

81
 Insignificant
INTR(-1) 0.002220 0.8250
Insignificant
MS 0.013668 0.8065
Significant
MS(-1) -0.086118 0.0414
Insignificant
EXGR 0.086455 0.3791
Insignificant
EXGR(-1) 0.083411 0.3598
Source: Author‟s Computation (2017) (See GMM result computed in table B of Appendix III)

From the table 4.5 above, it can be deduced that the lagged value of Gross Domestic

Product (GDP(-1)) and the lagged value of Money Supply (MS(-1)) exert significant effect on

economic growth in the UK.

In summary, the lagged value of Gross Domestic Product has significant effect on

economic growth of the three countries; meanwhile, the lagged value of money supply has

significant effect on the economy of UK and USA while money supply has significant effect on

economic growth of the USA. However, inflation and the lagged value of exchange rate were

found to exert significant effect on economic growth of Nigeria. Also, it can be deduced that

monetary policy lag has negative effect on economic growth as the significant lagged value of

the monetary policy variables exert negative effect on economic growth.

4.4 Test for Stationarity of Variables (Unit Root Test)

As a result of the general assumption that time series data are often revealed to be

stationary, it is imperative to conduct the Augmented Dickey Fuller Unit Root Test (ADF URT

test) in a bid to ensure stationarity of data. Also, the test is carried out to avoid problem of

82
spurious regression. The test for the stationarity of the variables in the research model is based

on the following hypothesis:

H0:- Xt has a unit root i.e. data is non-stationary

H1:- Xt has no unit root i.e. data is stationary

Decision Rule:

The ADF test statistics must be greater than 5% Mackinnon Critical Value in absolute

terms i.e. by ignoring the negativity of both the ADF test statistics an the Mackinnon critical

value before the variable can be adjudged as stationary by accepting the alternative hypothesis

and rejecting the null hypothesis, otherwise, the alternative hypothesis (H1) is rejected and the

null hypothesis (H0) is accepted. The Augmented Dickey Fuller Unit Root test as duly presented

in table D in the appendix is summarized in table 4.1 and 4.2 below:

83
4.4.1 Unit Root Test for Nigeria

Table 4.7: Result of ADF Unit Root Test at Level for Nigeria

Mackinnon
ADF Statistics
Variables Critical Value @ H0 H1 Remarks
Value
5%

GDP 0.586226 -2.963972 Accept Reject Non-Stationary

MPR -2.615724 -2.963972 Accept Reject Non-Stationary

INFL -2.966876 -2.963972 Reject Accept Stationary

INTR -1.762994 -2.963972 Accept Reject Non- Stationary

MS -1.646224 -2.971853 Accept Reject Non-Stationary

EXGR -3.281138 -2.963972 Reject Accept Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix II)
The table above revealed that of all the variables under consideration, only Inflation Rate

and Exchange Rate was found to be stationary before its first differencing as their ADF statistics

value was higher than Mackinnon Critical Value at 5%, hence, for the variables; INFL and

EXGR, the null hypothesis is rejected while the alternative hypothesis is accepted. However,

since other variables were found to be non-stationary at level, there is need to proceed to first

differencing to achieve stationarity of the variables in the case of Nigeria. Hence, the result of the

first differencing as duly presented in table D of appendix II is also summarized below:

84
Table 4.8: Result of ADF Unit Root Test at First Difference for Nigeria

Mackinnon
ADF Statistics
Variables Critical Value H0 H1 Remarks
Value
@ 5%

GDP -5.243327 -2.967767 Reject Accept Stationary

MPR -6.642865 -2.967767 Reject Accept Stationary

INTR -3.774111 -2.981038 Reject Accept Stationary

MS -3.284855 -2.971853 Reject Accept Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix II)
From the table above, it is revealed that all other variables (GDP, MPR, INTR and MS)

were found to be stationary at first difference as a result of the ADF statistics which is greater

than Mackinnon critical value at 5% in absolute terms. Hence, the null hypothesis is rejected for

the variables while the alternative hypothesis is accepted for the variables for the Nigerian case.

85
4.4.2 Unit Root Test for USA

Table 4.9: Result of ADF Unit Root Test at Level for USA

Mackinnon
ADF Statistics
Variables Critical Value @ H0 H1 Remarks
Value
5%

GDP -2.322079 -2.963972 Accept Reject Non-Stationary

MPR -1.135448 -2.971853 Accept Reject Non-Stationary

INFL -0.303452 -2.981038 Accept Reject Non-Stationary

INTR -1.949527 -2.967767 Accept Reject Non- Stationary

MS -0.243658 -2.967767 Accept Reject Non-Stationary

EXGR -2.858546 -2.976263 Accept Reject Non-Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix III)
The table above revealed that of all the variables under consideration, no variable was

found to be stationary before its first and second differencing as their ADF statistics value was

lower than Mackinnon Critical Value at 5%. However, since other variables were found to be

non-stationary at level, there is need to proceed to first differencing to achieve stationarity of the

variables in the case of USA. Hence, the result of the first differencing as duly presented in table

D of appendix III is also summarized below:

86
Table 4.10: Result of ADF Unit Root Test at First Difference for USA

Mackinnon
ADF Statistics
Variables Critical Value H0 H1 Remarks
Value
@ 5%

GDP -3.207701 -2.967767 Reject Accept Stationary

MPR -5.082998 -2.971853 Reject Accept Stationary

INFL -3.685236 -2.991878 Reject Accept Stationary

INTR -3.815764 -2.967767 Reject Accept Stationary

MS -2.823405 -2.967767 Accept Reject Non-Stationary

EXGR -4.297672 -2.967767 Reject Accept Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix III)
From the table above, it is revealed that all variables (GDP, MPR, INFL, INTR and

EXGR) except Money Supply (MS) were found to be stationary at first difference as a result of

the ADF statistics which is greater than Mackinnon critical value at 5% in absolute terms. Hence,

the null hypothesis is rejected for the variables while the alternative hypothesis is accepted for

the variables for the American case.

87
Table 4.11: Result of ADF Unit Root Test at Second Difference for USA

Mackinnon
ADF Statistics
Variables Critical Value H0 H1 Remarks
Value
@ 5%

MS -6.416142 -2.971853 Reject Accept Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix III)
From the table above, it is revealed that Money Supply (MS) was found to be stationary

at second difference as a result of the ADF statistics which is greater than Mackinnon critical

value at 5% in absolute terms. Hence, the null hypothesis is rejected for the variable while the

alternative hypothesis is accepted for the variables for the case of USA.

4.4.3 Unit Root Test for UK

Table 4.12: Result of ADF Unit Root Test at Level for UK

Mackinnon
ADF Statistics
Variables Critical Value @ H0 H1 Remarks
Value
5%

GDP -1.801552 -2.963972 Accept Reject Non-Stationary

MPR -0.244817 -2.971853 Accept Reject Non-Stationary

INFL -2.720764 -2.976263 Accept Reject Non-Stationary

INTR -2.183440 -3.020686 Accept Reject Non- Stationary

MS -0.603857 -2.967767 Accept Reject Non-Stationary

EXGR -2.841615 -2.967767 Accept Reject Non-Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix IV)

88
 The table above revealed that of all the variables under consideration, no variable was

found to be stationary before its first differencing as their ADF statistics value was lower than

Mackinnon Critical Value at 5%. However, since other variables were found to be non-stationary

at level, there is need to proceed to first differencing to achieve stationarity of the variables in the

case of UK. Hence, the result of the first differencing as duly presented in table D of appendix IV

is also summarized below:

Table 4.13: Result of ADF Unit Root Test at First Difference for UK

Mackinnon
ADF Statistics
Variables Critical Value H0 H1 Remarks
Value
@ 5%

GDP -3.295607 -2.967767 Reject Accept Stationary

MPR -4.368365 -2.976263 Reject Accept Stationary

INFL -6.753291 -2.981038 Reject Accept Stationary

INTR -3.886822 -3.212696 Reject Accept Stationary

MS -5.870081 -2.967767 Reject Accept Stationary

EXGR -4.106185 -2.967767 Reject Accept Stationary

Source: - Author‟s Computation (2017) (See URT results @ level in Table D of Appendix IV)
From the table above, it is revealed that all variables (GDP, MPR, INFL, INTR, MS and

EXGR) were found to be stationary at first difference as a result of the ADF statistics which is

greater than Mackinnon critical value at 5% in absolute terms. Hence, the null hypothesis is

rejected for the variables while the alternative hypothesis is accepted for the variables for the

case of UK.

89
4.5 Summary of Order of Co-Integration

The summary of the Augmented Dickey Fuller (ADF) unit root test is presented in the

tables below:

Table 4.14: - Summary of Order of Integration (Nigeria)

Variables Order of Integration

I(1)
GDP
I(1)
MPR

INFL I(0)

I(1)
INTR
I(1)
MS

EXGR I(0)

Source: - Author’s Compilation (2017)

Table 4.15: - Summary of Order of Integration (USA)

Variables Order of Integration

I(1)
GDP
I(1)
MPR
I(1)
INFL
I(1)
INTR

MS I(2)

I(1)
EXGR
Source: - Author’s Compilation (2017)

90
 Table 4.16: - Summary of Order of Integration (UK)

Variables Order of Integration

I(1)
GDP
I(1)
MPR
I(1)
INFL
I(1)
INTR
I(1)
MS
I(1)
EXGR
Source: - Author’s Compilation (2017)

From the tables above, it was discovered that all variables except inflation and exchange

rate which were stationary at level were stationary at first difference in Nigeria while all

variables were stationary at first difference in case of USA except Money Supply which was

stationary at first difference and all variables were stationary at first difference in the case of UK.

Hence, considering the mixed order of integration, it is necessary to proceed to the Auto

Regressive Distributed Lag (ARDL) model to examine the long run relationship among the

variables rather than the co-integration test which should only be used when variables are co-

integrated in same order.

4.6 ARDL Bounds Test Approach to Co-Integration

The bounds testing approach as developed by Pesaran, Shin and Smith (2001) is

employed to test for co-integration within the ARDL framework, this test to test for a long run

equilibrium relationship between the variables. The test is based on the following hypothesis:

Ho: There is no co-integration among variables

H1: There is co-integration among variables

91
Decision Rule:

The F-Statistics of the model must be greater than the upper bound of the test result at 5%

significant level for a co-integrating relationship to exist. Hence, when the F-Statistics is greater

than upper bound at 5% significant level, the alternative hypothesis which assumes that there

exists co-integration among variables is accepted which shows that there is long run relationship

among variables and if otherwise, the null hypothesis is accepted. Hence, the co-integration

result is presented in table F of appendix II and summarized in table 4.5 below:

4.6.1 Nigeria

The study adopted the Akaike Information Criterion (AIC) for the selection of the ARDL

(1, 0, 0, 0, 0, 0) model in case of Nigeria.

Table 4.17: Co-Integration Result (Nigeria)

F-Statistics Lower Bound (5%) Upper Bound (5%)

4.520553 2.04 3.24

Source: Author‟s Computation (2017) (See Bound Test result computed in table F of Appendix II)

Therefore, considering the results specified above, it can be deduced that there exists a

stable long run equilibrium relationship among variables as the alternative hypothesis is accepted

because the F-Statistics was found to be greater than upper bound at 5% critical value in the case

of Nigeria.

4.6.2 USA

The study adopted the Akaike Information Criterion (AIC) for the selection of the ARDL

(1, 0, 0, 0, 0, 0) model in case of USA.

92
Table 4.18: Co-Integration Result (USA)

F-Statistics Lower Bound (5%) Upper Bound (5%)

14.29118 2.04 3.24

Source: Author‟s Computation (2017) (See Bound Test result computed in table F of Appendix III)

Therefore, considering the results specified above, it can be deduced that there exists a

stable long run equilibrium relationship among variables as the alternative hypothesis is accepted

because the F-Statistics was found to be greater than upper bound at 5% critical value in the case

of USA.

4.6.3 UK

The study adopted the Akaike Information Criterion (AIC) for the selection of the ARDL

(1, 0, 0, 0, 0, 0) model in case of UK.

Table 4.19: Co-Integration Result (UK)

F-Statistics Lower Bound (5%) Upper Bound (5%)

18.51191 2.04 3.24

Source: Author‟s Computation (2017) (See Bound Test result computed in table F of Appendix IV)

Therefore, considering the results specified above, it can be deduced that there exists a

stable long run equilibrium relationship among variables as the alternative hypothesis is accepted

because the F-Statistics was found to be greater than upper bound at 5% critical value in the case

of UK.

Hence, it can be deduced that there exists a long run relationship between monetary

policy and economic growth in the three countries.

93
4.7 Long-Run Results

The long run result of the model for the three countries as obtained through the use of the

ARDL technique as presented in table E in appendix II, III and IV is summarized below in table

4.6:

4.7.1 Nigeria

Table 4.20: Long Run Result of the Model (Nigeria)

Dependent Variable: GDP

Variable Co-Efficient Std. Error T-Statistics Prob.

MPR -0.128198 0.333066 -0.384904 0.7038

INFL -0.021715 0.130889 -0.165903 0.8697
INTR 0.185392 0.413274 0.448594 0.6579
MS 0.290865 0.083859 3.468484 0.0021
EXGR -0.170319 0.257754 -0.660779 0.5153
C 18.964158 3.767020 5.034260 0.0000
Source: Author‟s Computation (2017) (See Bound Test result computed in table E of Appendix II)

From the table above, the long run equation specifying the long run relationship among

the variables for the case of Nigeria can be presented below as:

GDP = 18.964158 – 0.128198MPR – 0.021715INFL + 0.185392INTR + 0.290865MS – 0.170319EXGR + µ

(3.767020) (0.333066) (0.130889) (0.413274) (0.083859) (0.257754)

Note: The Standard Error Statistics are those stated in parenthesis

From the long run equation above, the coefficient of the constant parameter was found to

be 18.964158 units which means that if all variables are held constant in the long run, GDP

which is the explained variable will increase by 18.964158 units. Also, Monetary Policy Rate

(MPR), Inflation Rate (INFL) and Exchange Rate (EXGR) were found to be negatively related to

GDP to the tune of -0.128198, -0.021715 and -0.170319 units respectively which means that a

unit increase in Monetary Policy Rate, Inflation Rate and Exchange Rate will reduce GDP by

94
same units in the long run. However, Interest Rate (INTR) and Money Supply (MS) were found

to be positively related to GDP by 0.185392 and 0.290865 units respectively which implies that a

unit increase in INTR and MS will increase GDP by same units in Nigeria.

4.7.2 USA

Table 4.21: Long Run Result of the Model (USA)

Dependent Variable: GDP

Variable Co-Efficient Std. Error T-Statistics Prob.

MPR -0.415770 0.783071 -0.530948 0.6008

INFL 0.191541 0.249884 0.766521 0.4515
INTR 0.231184 0.447721 0.516357 0.6108
MS 0.736585 0.360925 2.040829 0.0534
EXGR -0.000182 0.834883 -0.000218 0.9998
C 7.959018 10.383005 0.766543 0.4515
Source: Author‟s Computation (2017) (See Bound Test result computed in table E of Appendix III)

From the table above, the long run equation specifying the long run relationship among

the variables for the case of USA can be presented below as:

GDP = 7.959018 – 0.415770MPR + 0.191541INFL + 0.231184INTR + 0.736585MS – 0.000182EXGR + µ

(10.383005) (0.783071) (0.249884) (0.447721) (0.360925) (0.834883)

Note: The Standard Error Statistics are those stated in parenthesis

From the long run equation above, the coefficient of the constant parameter was found to

be 7.959018 units which means that if all variables are held constant in the long run, GDP which

is the explained variable will increase by 7.959018 units. Also, Monetary Policy Rate (MPR) and

Exchange Rate were found to be negatively related to GDP to the tune of -0.415770 and

-0.000182 units which means that a unit increase in Monetary Policy Rate and Exchange Rate

will reduce GDP by same units in the long run. However, Inflation (INFL), Interest Rate (INTR),

and Money Supply (MS) were found to be positively related to GDP by 0.191541, 0.231184 and

95
0.736585 units respectively which implies that a unit increase in INFL, INTR and MS will

increase GDP by same units in USA.

4.7.3 United Kingdom

Table 4.22: Long Run Result of the Model (UK)

Dependent Variable: GDP

Variable Co-Efficient Std. Error T-Statistics Prob.

MPR -0.454971 1.148627 -0.396099 0.6990

INFL 0.534722 1.597021 0.334825 0.7435
INTR 0.106834 0.401268 0.266241 0.7946
MS 0.231656 0.223696 1.035582 0.3208
EXGR 0.667035 1.942477 0.343394 0.7372
C 18.630471 11.668807 1.596605 0.1363
Source: Author‟s Computation (2017) (See Bound Test result computed in table E of Appendix IV)

From the table above, the long run equation specifying the long run relationship among

the variables for the case of UK can be presented below as:

GDP = 18.630471 – 0.454971MPR + 0.534722INFL + 0.106834INTR + 0.231656MS + 0.667035EXGR + µ

(11.668807) (1.148627) (1.597021) (0.401268) (0.223696) (1.942477)

Note: The Standard Error Statistics are those stated in parenthesis

From the long run equation above, the coefficient of the constant parameter was found to

be 18.630471 units which means that if all variables are held constant in the long run, GDP

which is the explained variable will increase by 18.630471 units. Also, Monetary Policy Rate

(MPR) was found to be negatively related to GDP to the tune of -0.454971 which means that a

unit increase in Monetary Policy Rate will reduce GDP by same units in the long run. However,

Inflation (INFL), Interest Rate (INTR), Money Supply (MS) and Exchange Rate (EXGR) were

found to be positively related to GDP by 0.534722, 0.106834, 0.231656 and 0.667035 units

96
respectively which implies that a unit increase in INFL, INTR, MS and EXGR will increase

GDP by same units in UK.

Hence, from the long run results of the countries, it can be inferred that monetary policy

rate, inflation rate and exchange rate exert negative effect on economic growth in Nigeria as only

monetary policy rate has negative effect on economic growth in the UK while monetary policy

rate and exchange rate has negative effect on economic growth in the USA as other variables

exert positive influence on economic growth in all countries.

4.8 Test for Statistical Significance of Parameters in the Long Run

(Probability Test)

In a bid to test for the statistical significance of the variables in the study, the probability

test will be employed for this purpose. This is done by considering the probability value attached

to the variables as presented in the ARDL long run results.

Decision Rule:

If the probability value attached to the variable is less than 0.05 i.e. 5% significant is

statistically significant but if otherwise, the variable is considered to be statistically insignificant.

4.8.1 Nigeria

Table 4.23: Long Run Probability Test (Nigeria)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Insignificant
MPR -0.128198 0.7038
Insignificant
INFL -0.021715 0.8697

97
 Insignificant
INTR 0.185392 0.6579
Significant
MS 0.290865 0.0021
Insignificant
EXGR -0.170319 0.5153
Source: Author‟s Computation (2017) (See ARDL result computed in table E of Appendix II)

From the table 4.23 above, it can be deduced that money supply was the only monetary

policy variable discovered to have significant effect on economic growth in Nigeria.

4.8.2 USA

Table 4.24: Probability Test (USA)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Insignificant
MPR -0.415770 0.6008
Insignificant
INFL 0.191541 0.4515
Insignificant
INTR 0.231184 0.6108
Insignificant
MS 0.736585 0.0534
Insignificant
EXGR -0.000182 0.9998
Source: Author‟s Computation (2017) (See ARDL result computed in table E of Appendix III)

From the table 4.23 above, it can be deduced that no monetary policy variable has

significant effect on economic growth in USA.

98
4.8.3 United Kingdom

Table 4.25: Probability Test (UK)

Dependent Variable: GDP

Independent Probability
Coefficient Decision Rule
Variables Value

Insignificant
MPR -0.454971 0.6990
Insignificant
INFL 0.534722 0.7435
Insignificant
INTR 0.106834 0.7946
Insignificant
MS 0.231656 0.3208
Insignificant
EXGR 0.667035 0.7372
Source: Author‟s Computation (2017) (See ARDL result computed in table E of Appendix IV)

From the table 4.23 above, it can be deduced that no monetary policy variable has

significant effect on economic growth in UK.

Hence, from table 4.23 to 4.25, it can be inferred that monetary policy has no significant

effect on economic growth in UK and USA while Money Supply has a substantial effect on

economic growth in Nigeria as a monetary policy variable.

4.9 Diagnostic and Stability Tests

Diagnostic and stability tests are the tests carried out to test for the robustness, stability

and reliability of the overall model through various techniques. The diagnostic test encompasses

the serial correlation or autocorrelation test, heteroskedaticity test, Ramsey RESET test and

CUSUM test.

99
4.9.1 Serial Correlation Test

In this study, the test for the serial or autocorrelation of residuals was based on the

Breusch-Godfrey Serial Correlation Langrange Multiplier (LM) test. The LM test is a general

test for error auto correlation (Asteriou & Hall, 2011). The test for the serial correlation in the

model adopted for the study is based on the following hypothesis:

H0:- There is no serial correlation

H1:- There is serial correlation

Decision Rule:

If the Probability Value (P-Value) of the F-Statistic is more than 5%, then there is no

auto correlation ad the null hypothesis is accepted but if otherwise, the alternative hypothesis is

accepted.

Table 4.26 Result of the Breusch-Godfrey Serial Correlation LM Test (Nigeria)

F-Statistics 0.338172 Prob. F(2,20) 0.7171

Obs* R-Squared 0.948605 Prob. Chi-Square (2) 0.6223

Source: Author‟s Computation using E-Views 9 (2017). (See table F of appendix II for result)

From table 4.26 above, the result shows that the value of the F-Statistics is 0.338172

while the P-Value is more than 5% at 0.7171. Hence, the null hypothesis of no auto-correlation is

accepted in case of Nigeria.

Table 4.27 Result of the Breusch-Godfrey Serial Correlation LM Test (USA)

F-Statistics 0.643029 Prob. F(2,18) 0.5374

Obs* R-Squared 1.800449 Prob. Chi-Square (2) 0.4065

Source: Author‟s Computation using E-Views 9 (2017). (See table F of appendix III for result)

100
 From table 4.27 above, the result shows that the value of the F-Statistics is 0.643029

while the P-Value is more than 5% at 0.5374. Hence, the null hypothesis of no auto-correlation is

accepted in case of USA.

Table 4.28 Result of the Breusch-Godfrey Serial Correlation LM Test (UK)

F-Statistics 0.207913 Prob. F(2,7) 0.8171

Obs* R-Squared 0.897166 Prob. Chi-Square (2) 0.6385

Source: Author‟s Computation using E-Views 9 (2017). (See table F of appendix IV for result)

From table 4.28 above, the result shows that the value of the F-Statistics is 0.207913

while the P-Value is more than 5% at 0.8171. Hence, the null hypothesis of no auto-correlation is

accepted in case of UK.

Hence, the results presented in table 4.26 to table 4.28 revealed that the model for the

three countries can be relied upon as a basis for making inferences and valid policy

recommendations as auto correlation was absent in the study.

4.9.2 Heteroskedasticity Test

Heteroskedasticity is a common problem in data analysis. It occurs when the variance of

errors varies across observations; hence, it is imperative to test for heteroskedaticity as the

estimated standard error can be either too large or too small. Hence, it can result in incorrect

inferences (Hendry, 1995). As a result, to test for heteroskedasticity, the test is based on the

following hypothesis:

H0:- There is no heteroskedasticity in the model

H1:- There is heteroskedasticity in the model

101
Decision Rule:

If the probability value of the F-Statistics is more than 5%, there is no heteroskedasticity

in the study and the null hypothesis is accepted but if otherwise, the alternative hypothesis is

accepted.

Table 4.29 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (Nigeria)

F-statistic 0.435617 Prob. F(7,21) 0.8687

Obs*R-squared 3.677035 Prob. Chi-Square(7) 0.8161
Scaled explained SS 11.50814 Prob. Chi-Square(7) 0.1179
Source: Author‟s Computation using E-Views 9 (2017). (See table G of appendix II for result)

From table 4.29 above, the White Heteroskedasticity test has an F-Statistics of 0.435617

and Probability Value is more than 5% at 0.0517. Hence, the null hypothesis of no

heteroskedasticity is accepted and it can be deduced that the model has no heteroskedasticity

problem as regards the case of Nigeria.

Table 4.30 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (USA)

F-statistic 0.280285 Prob. F(7,19) 0.9540

Obs*R-squared 2.527135 Prob. Chi-Square(7) 0.9250
Scaled explained SS 2.202701 Prob. Chi-Square(7) 0.9478
Source: Author‟s Computation using E-Views 9 (2017). (See table G of appendix III for result)

From table 4.30 above, the White Heteroskedasticity test has an F-Statistics of 0.280285

and Probability Value is more than 5% at 0.9540. Hence, the null hypothesis of no

heteroskedasticity is accepted and it can be deduced that the model has no heteroskedasticity

problem as regards the case of USA.

Table 4.31 Breusch-Pagan-Godfrey Heteroskedasticity Test Result (UK)

F-statistic 1.423832 Prob. F(7,8) 0.3141

Obs*R-squared 8.875758 Prob. Chi-Square(7) 0.2617
Scaled explained SS 0.727097 Prob. Chi-Square(7) 0.9981
Source: Author‟s Computation using E-Views 9 (2017). (See table G of appendix IV for result)

102
 From table 4.31 above, the White Heteroskedasticity test has an F-Statistics of 1.423832

and Probability Value is more than 5% at 0.3141. Hence, the null hypothesis of no

heteroskedasticity is accepted and it can be deduced that the model has no heteroskedasticity

problem as regards the case of UK.

Hence, the Heteroskedasticity test for the three countries revealed that the model for the

three countries is devoid of the problem of Heteroskedasticity.

4.9.3 Ramsey RESET Test (Functionality Test)

The test is carried out to test the functionality relationship between variables, that is, if

the explanatory variables can sufficiently explain the dependent variable. The test is carried out

based on the following hypothesis:

H0: There is no problem of functionality in the model

H1: There is problem of functionality in the model

Decision Rule:

If the probability value of the F-Statistic is more than 5%, then the model has no problem

of functionality and the null hypothesis is accepted while the alternative hypothesis is rejected.

Table 4.32 Ramsey RESET Test Result (Nigeria)

Ramsey RESET 3.074015 Probability Value 0.0941

F-Statistics

Source: Author‟s Computation using E-Views 9 (2017). (See table H of appendix II for result)

From the table 4.32 above, it is showed by the probability value of the Ramsey RESET

test at 0.0941 which is more than 5% that the model has no problem of functionality as the null

hypothesis is accepted and the alternative hypothesis is rejected for Nigeria.

103
Table 4.33 Ramsey RESET Test Result (USA)

Ramsey RESET 0.993516 Probability Value 0.3329

F-Statistics

Source: Author‟s Computation using E-Views 9 (2017). (See table H of appendix III for result)

From the table 4.33 above, it is showed by the probability value of the Ramsey RESET

test at 0.3329 which is more than 5% that the model has no problem of functionality as the null

hypothesis is accepted and the alternative hypothesis is rejected for USA.

Table 4.34 Ramsey RESET Test Result (UK)

Ramsey RESET 0.423801 Probability Value 0.6829

F-Statistics

Source: Author‟s Computation using E-Views 9 (2017). (See table H of appendix IV for result)

From the table 4.34 above, it is showed by the probability value of the Ramsey RESET

test at 0.6829 which is more than 5% that the model has no problem of functionality as the null

hypothesis is accepted and the alternative hypothesis is rejected for UK.

Hence, the table 4.32 to 4.34 revealed that the study has no problem of functionality for

the three countries.

4.9.4 Stability Test - Test for Structural Break (CUSUM Test)

The test is carried out to assess stability of the model, for this purpose, the Cumulative

Sum of Recursive Residuals (CUSUM) test is applied to assess the parameter stability (Pesaran

& Pesaran, 1997). The test is formulated based on the following hypothesis:

H0: There is no presence of stability in the model

H1: There is presence of stability in the model

104
Decision Rule:

If the plot of CUSUM statistic line falls inside the critical bands of the 5% confidence

interval of parameter stability, then, there is presence of stability and the alternative hypothesis is

accepted while the null hypothesis is rejected.

Fig. 4.1: Cumulative Sum of Squares (CUSUM) Test (Nigeria)

15

10

-5

-10

-15
94 96 98 00 02 04 06 08 10 12 14 16

CUSUM 5% Significance
Source: Output of Analysis through E-Views 9 (2017) (See Table I of appendix II).
From the figure above, it is discovered that the CUSUM line graph falls in between the

5% critical bands which means that there is no problem of instability in the model as structural

break is absent, hence, the alternative hypothesis is accepted and the null hypothesis is rejected

for Nigeria.

105
Fig. 4.2: Cumulative Sum of Squares (CUSUM) Test (USA)
15

10

-5

-10

-15
94 96 98 00 02 04 06 08 10 12 14 16

CUSUM 5% Signific anc e

Source: Output of Analysis through E-Views 9 (2017) (See Table I of appendix III).

From the figure above, it is discovered that the CUSUM line graph falls in between the 5%

critical bands which means that there is no problem of instability in the model as structural break

is absent, hence, the alternative hypothesis is accepted and the null hypothesis is rejected for

USA.

Fig. 4.3: Cumulative Sum of Squares (CUSUM) Test (UK)

12

-4

-8

-12
96 97 98 99 00 01 02 03 04 05 06 07 08

CUS UM 5% S ignific anc e

Source: Output of Analysis through E-Views 9 (2017) (See Table I of appendix III).

From the figure above, it is discovered that the CUSUM line graph falls in between the 5%

critical bands which means that there is no problem of instability in the model as structural break

is absent, hence, the alternative hypothesis is accepted and the null hypothesis is rejected for UK.
106
 Hence, the fig. 4.1 to 4.3 revealed that the study has no problem of instability for the

three countries.

4.10 Interpretation of the Granger Causality Test (G. C. Test)

The Granger Causality test is usually carried out to determine the causation that exists

between variables in a research model. It was specified in the early chapter of this study as an

objective, to determine the causal relations between the variables to be used in the study.

However, one major shortcoming of the co-integration test which necessitated for the need for

causality tests is that, it can only show the existence of a long run relationship; hence it fails in

the establishment of causal relations between variables.

Decision Rule: -

If the probability value attached to a variable is less than 10% and its F-calculated is

greater than the F-tabulated at 95% confidence level (5% significant level), we accept the

Alternate Hypothesis (H1) and reject Null Hypothesis (H0), but if the probability value that is

attached to a variable is greater than 10% and its F-calculated is less than the F-tabulated at 95%

confidence level (5% significant level), we accept the Null Hypothesis (H0) and reject the

Alternate Hypothesis (H1).

107
4.10.1 Nigeria

4.10.1.1 Hypothesis One

Table 4.35 Monetary Policy Rate (MPR) and Gross Domestic Product (GDP)-Nigeria

HYPOTHESIS F-Statistics Probability

A H0: - MPR does not Granger Cause GDP

0.15150 0.8602
H1: - MPR does Granger Cause GDP

B H0: - GDP does not Granger Cause MPR

1.51394 0.2403
H1: - GDP does Granger Cause MPR

Source: Author‟s Computation (2017) (See Granger Causality test result in table J in appendix II)

From the table 4.35 above, it is revealed that there exists no causality between Monetary

Policy Rate (MPR) and Gross Domestic Product (GDP) as the null hypothesis is accepted in both

cases.

4.10.1.2 Hypothesis Two

Table 4.36 Inflation (INFL) and Gross Domestic Product (GDP)- Nigeria

HYPOTHESIS F-Statistics Probability

A H0: - INFL does not Granger Cause GDP

0.19414 0.8248
H1: - INFL does Granger Cause GDP

B H0: - GDP does not Granger Cause INFL

2.28891 0.1231
H1: - GDP does Granger Cause INFL

Source: Author‟s Computation (2017) (See Granger Causality test result in table J in appendix II)

From the table above, it can be deduced that there exists no causality between Inflation (INFL)

and Gross Domestic Product (GDP) as the null hypothesis is accepted in both cases.

108
4.10.1.3 Hypothesis Three

Table 4.37 Interest Rate (INTR) and Gross Domestic Product (GDP) - Nigeria

HYPOTHESIS F-Statistics Probability

A H0: - INTR does not Granger Cause GDP

0.25095 0.7801
H1: - INTR does Granger Cause GDP

B H0: - GDP does not Granger Cause INTR

3.62778 0.0420
H1: - GDP does Granger Cause INTR

Source: Author‟s Computation (2017) (See Granger Causality test result in table J in appendix II)

From the table above, it can be deduced that there exists a unidirectional causality running from

Gross Domestic Product (GDP) to Interest Rate (INTR) as the null hypothesis is accepted in case

A and rejected in B.

4.10.1.4 Hypothesis Four

Table 4.38 Money Supply (MS) and Gross Domestic Product (GDP)- Nigeria

HYPOTHESIS F-Statistics Probability

A H0: - MS does not Granger Cause GDP

1.32348 0.2849
H1: - MS does Granger Cause GDP

B H0: - GDP does not Granger Cause MS

0.48099 0.6240
H1: - GDP does Granger Cause MS

Source: Author‟s Computation (2017) (See Granger Causality test result in table J in appendix II)

From the table above, it can be deduced that there exists no causality between Gross Domestic

Product (GDP) and Money Supply (MS) as the null hypothesis is accepted in both cases.

109
4.10.1.5 Hypothesis Five

Table 4.39 Exchange Rate (EXGR) and Gross Domestic Product (GDP) - Nigeria

HYPOTHESIS F-Statistics Probability

A H0: - EXGR does not Granger Cause GDP

0.61308 0.5499
H1: - EXGR does Granger Cause GDP

B H0: -GDP does not Granger Cause EXGR

0.46142 0.6359
H1: - GDP does Granger Cause EXGR

Source: Author‟s Computation (2017) (See Granger Causality test result in table J in appendix II)

From the table 4.39 above, it can be extrapolated that there exists no causality between

Exchange Rate (EXGR) and Gross Domestic Product (GDP) as the null hypothesis is accepted in

both cases.

Hence, considering the results presented in table 4.35 to table 4.39, it was revealed that

Gross Domestic Product (GDP) causes Interest Rate (INTR) while there exists no causality

between each of Monetary Policy Rate (MPR), Inflation Rate (INFL), Money Supply (MS),

Exchange Rate (EXGR) and Gross Domestic Product (GDP) in the Nigerian context.

110
4.10.2 USA

4.10.2.1 Hypothesis One

Table 4.40 Monetary Policy Rate (MPR) and Gross Domestic Product (GDP) - USA

HYPOTHESIS F-Statistics Probability

A H0: - MPR does not Granger Cause GDP

2.19591 0.1331
H1: - MPR does Granger Cause GDP

B H0: - GDP does not Granger Cause MPR

10.3340 0.0006
H1: - GDP does Granger Cause MPR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table 4.40 above, it is revealed that there exists a unidirectional causality

running from Monetary Policy Rate (MPR) to Gross Domestic Product (GDP) as the null

hypothesis is rejected in case B and accepted in case A.

4.10.2.2 Hypothesis Two

Table 4.41 Inflation (INFL) and Gross Domestic Product (GDP)- USA

HYPOTHESIS F-Statistics Probability

A H0: - INFL does not Granger Cause GDP

2.53237 0.1035
H1: - INFL does Granger Cause GDP

B H0: - GDP does not Granger Cause INFL

1.74106 0.1997
H1: - GDP does Granger Cause INFL

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix III)

From the table above, it can be deduced that there exists no causality between Inflation (INFL)

and Gross Domestic Product (GDP) as the null hypothesis is accepted in both cases.

111
4.10.2.3 Hypothesis Three

Table 4.42 Interest Rate (INTR) and Gross Domestic Product (GDP) - USA

HYPOTHESIS F-Statistics Probability

A H0: - INTR does not Granger Cause GDP

0.82085 0.4520
H1: - INTR does Granger Cause GDP

B H0: - GDP does not Granger Cause INTR

5.51882 0.0107
H1: - GDP does Granger Cause INTR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix III)

From the table above, it can be deduced that there exists a unidirectional causality running from

Gross Domestic Product (GDP) to Interest Rate (INTR) as the null hypothesis is accepted in case

A and rejected in B.

4.10.2.4 Hypothesis Four

Table 4.43 Money Supply (MS) and Gross Domestic Product (GDP) - USA

HYPOTHESIS F-Statistics Probability

A H0: - MS does not Granger Cause GDP

0.29109 0.7501
H1: - MS does Granger Cause GDP

B H0: - GDP does not Granger Cause MS

18.2689 2.E-05
H1: - GDP does Granger Cause MS

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix III)

From the table above, it can be deduced that there exists no causality between Gross Domestic

Product (GDP) and Money Supply (MS) as the null hypothesis is accepted in both cases.

112
4.10.2.5 Hypothesis Five

Table 4.44 Exchange Rate (EXGR) and Gross Domestic Product (GDP) - USA

HYPOTHESIS F-Statistics Probability

A H0: - EXGR does not Granger Cause GDP

0.54805 0.5851
H1: - EXGR does Granger Cause GDP

B H0: -GDP does not Granger Cause EXGR

1.92416 0.1679
H1: - GDP does Granger Cause EXGR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix III)

From the table 4.39 above, it can be extrapolated that there exists no causality between

Exchange Rate (EXGR) and Gross Domestic Product (GDP) as the null hypothesis is accepted in

both cases.

Hence, considering the results presented in table 4.36 to table 4.44, it was revealed that

Gross Domestic Product (GDP) causes Monetary Policy Rate (MPR) and Interest Rate (INTR)

while there exists no causality between each of Inflation Rate (INFL), Money Supply (MS),

Exchange Rate (EXGR) and Gross Domestic Product (GDP) in the USA context.

113
4.10.3 United Kingdom

4.10.3.1 Hypothesis One

Table 4.45 Monetary Policy Rate (MPR) and Gross Domestic Product (GDP) - UK

HYPOTHESIS F-Statistics Probability

A H0: - MPR does not Granger Cause GDP

8.02204 0.0024
H1: - MPR does Granger Cause GDP

B H0: - GDP does not Granger Cause MPR

6.84234 0.0049
H1: - GDP does Granger Cause MPR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table 4.40 above, it is revealed that there exists a bi-directional causality

between Monetary Policy Rate (MPR) and Gross Domestic Product (GDP) as the null hypothesis

is rejected in both cases.

4.10.3.2 Hypothesis Two

Table 4.46 Inflation (INFL) and Gross Domestic Product (GDP) - UK

HYPOTHESIS F-Statistics Probability

A H0: - INFL does not Granger Cause GDP

0.85814 0.4383
H1: - INFL does Granger Cause GDP

B H0: - GDP does not Granger Cause INFL

1.58696 0.2281
H1: - GDP does Granger Cause INFL

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table above, it can be deduced that there exists no causality between Inflation (INFL)

and Gross Domestic Product (GDP) as the null hypothesis is accepted in both cases.

114
4.10.3.3 Hypothesis Three

Table 4.47 Interest Rate (INTR) and Gross Domestic Product (GDP) - UK

HYPOTHESIS F-Statistics Probability

A H0: - INTR does not Granger Cause GDP

1.46729 0.2663
H1: - INTR does Granger Cause GDP

B H0: - GDP does not Granger Cause INTR

3.04419 0.0823
H1: - GDP does Granger Cause INTR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table above, it can be deduced that there exists a unidirectional causality running from

Gross Domestic Product (GDP) to Interest Rate (INTR) as the null hypothesis is accepted in case

A and rejected in B.

4.10.3.4 Hypothesis Four

Table 4.48 Money Supply (MS) and Gross Domestic Product (GDP) - UK

HYPOTHESIS F-Statistics Probability

A H0: - MS does not Granger Cause GDP

0.38526 0.6844
H1: - MS does Granger Cause GDP

B H0: - GDP does not Granger Cause MS

9.17098 0.0011
H1: - GDP does Granger Cause MS

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table above, it can be deduced that there exists a unidirectional causality running from

Gross Domestic Product (GDP) to Money Supply (MS) as the null hypothesis is accepted in case

A and rejected in B.

115
4.10.3.5 Hypothesis Five

Table 4.49 Exchange Rate (EXGR) and Gross Domestic Product (GDP) - UK

HYPOTHESIS F-Statistics Probability

A H0: - EXGR does not Granger Cause GDP

0.67054 0.5208
H1: - EXGR does Granger Cause GDP

B H0: -GDP does not Granger Cause EXGR

0.92691 0.4095
H1: - GDP does Granger Cause EXGR

Source: Author‟s Computation (2017) (See Granger Causality test result in table K in appendix IV)

From the table 4.39 above, it can be extrapolated that there exists no causality between

Exchange Rate (EXGR) and Gross Domestic Product (GDP) as the null hypothesis is accepted in

both cases.

Hence, considering the results presented in table 4.44 to table 4.49, it was revealed that

there exists a bi-directional causality between Gross Domestic Product (GDP) and Monetary

Policy Rate (MPR) and Gross Domestic Product (GDP) causes Interest Rate (INTR) and Money

Supply (MS) while there exists no causality between each of Inflation Rate (INFL), Exchange

Rate (EXGR) and Gross Domestic Product (GDP) in the UK context.

Summarily, considering the Engle Granger causality result for the three countries, it was

discovered that economic growth causes Interest Rate while there exists no causality between

each of Monetary Policy Rate, Inflation Rate, Money Supply, Exchange Rate and economic

growth in the Nigerian context as compared to the USA context where economic growth causes

Monetary Policy Rate and Interest Rate while there exists no causality between each of Inflation

Rate, Money Supply, Exchange Rate and economic growth. Also, in UK, it was revealed that

there exists a bi-directional causality between economic growth and Monetary Policy Rate and

116
economic growth causes Interest Rate and Money Supply while there exists no causality between

each of Inflation Rate, Exchange Rate and economic growth in the UK context.

4.11 Summary of the Research Findings

The objective of the study is to empirically investigate the impact of monetary policy and

its lag on economic growth taking into cognisance a comparative study of Nigeria, United States

of America (USA) and United Kingdom (UK). The test for the stationarity of the variables was

carried out using the Augmented Dickey Fuller Unit Root Test revealing that all variables except

inflation and exchange rate which were stationary at level were stationary at first difference in

Nigeria while all variables were stationary at first difference in case of USA except Money

Supply which was stationary at first difference and all variables were stationary at first difference

in the case of UK. Hence, the mixed integration of the variables at different stationarity points

necessitated the need for the use of the Auto Regressive Distributed Lag (ARDL) method to

estimate the long run equilibrium relationship among variables. Meanwhile, the ARDL Bounds

testing approach to co-integration revealed that there exists a stable long run relationship among

the variables in all the countries.

Evidence from the short run model carried out through the Generalized Methods of

Moments (GMM) technique to examine the effect of monetary lag on economic growth revealed

that of all the variables considered; the lagged value of Gross Domestic Product has significant

effect on economic growth of the three countries; meanwhile, the lagged value of money supply

has significant effect on the economy of UK and USA while money supply has significant effect

on economic growth of the USA. However, inflation and the lagged value of exchange rate were

found to exert significant effect on economic growth of Nigeria. However, it can be deduced that

monetary policy rate, its lagged value and inflation has negative effect on economic growth in

117
the three countries. However, Money supply and its lagged value had a negative and positive

effect respectively on economic growth in Nigeria while money supply lag had negative effect

on economic growth in USA and UK while money supply has positive effect on economic

growth in both countries.

On the other hand, the long run model carried out through the ARDL methodology

revealed that monetary policy rate, inflation rate and exchange rate exert negative effect on

economic growth in Nigeria as only monetary policy rate has negative effect on economic

growth in the UK while monetary policy rate and exchange rate has negative effect on economic

growth in the USA as other variables exert positive influence on economic growth in all

countries in the long run. However, monetary policy has no significant effect on economic

growth in UK and USA while Money Supply has a substantial effect on economic growth in

Nigeria as a monetary policy variable. Meanwhile, all variables were found to conform to the a-

priori expectation in the long run.

Furthermore, in line with the specific objective of this study which is to determine the

causal relationship between variables used in the study as well as the direction of causality, the

study employed the Engle Granger Causality Technique for this purpose. Consequently, it was

discovered that economic growth causes Interest Rate while there exists no causality between

each of Monetary Policy Rate, Inflation Rate, Money Supply, Exchange Rate and economic

growth in the Nigerian context as compared to the USA context where economic growth causes

Monetary Policy Rate and Interest Rate while there exists no causality between each of Inflation

Rate, Money Supply, Exchange Rate and economic growth. Also, in UK, it was revealed that

there exists a bi-directional causality between economic growth and Monetary Policy Rate and

118
economic growth causes Interest Rate and Money Supply while there exists no causality between

each of Inflation Rate, Exchange Rate and economic growth in the UK context.

In addition, in a bid to test for the reliability and robustness of the whole model,

diagnostic and stability tests were adopted in a bid to test for the presence of serial or auto

correlation, heteroskedasticity, functional relationship (functionality) between variables and

stability through the LM Serial Correlation test, Breusch-Pagan Godfrey Heteroskedasticity test,

Ramsey RESET test and Cumulative Sum of Squares (CUSUM) test respectively. The tests all

revealed that problem of autocorrelation, functionality, instability and heteroskedasticity is

absent in the study as pertaining to all countries. Hence, the findings of the study can be relied on

for proper recommendations and decision making.

4.12 Implication of Research Findings

The objective of this study is to examine the impact of monetary policy and its lag on

economic growth taking into cognisance a comparative study of Nigeria, USA and UK. The

Auto Regressive Distributed Lag (ARDL) model result revealed that monetary policy rate,

inflation rate and exchange rate exert negative effect on economic growth in Nigeria as only

monetary policy rate has negative effect on economic growth in the UK while monetary policy

rate and exchange rate has negative effect on economic growth in the USA as other variables

exert positive influence on economic growth in all countries in the long run. Meanwhile, all

variables were found to conform to the a-priori expectation.

In consonance with the a-priori expectation, Monetary Policy Rate was found to be

negatively related to economic growth in the long run in the three countries. This implies that

increase in the monetary rate will lead to a decline in economic growth in the UK, USA and

Nigeria. This is probable because the increase in monetary policy which can be considered as a

119
tight or contractionary monetary control will lead to banks increasing the rate at which credit is

extended to the economy at large. As a result, only very few people will borrow for investment

and other purposes which will slow down the rate of economic development. Also, Inflation was

adjudged to be negatively related to economic growth in Nigeria as compared to its positive

effect on economic growth in USA and UK in conformity with the a-priori expectation which

implies that the an increase in the inflation rate in Nigeria will reduce economic growth and spur

growth in UK and USA. This is possible because a level of inflation is required to spur economic

growth and keep funds in check in the economy, hence, in so much as the two developed

countries (UK and USA) have kept a very low inflation rate as between the range of 1% and 4%,

an increase in such rate still tends to spur economic growth. However, in comparison to Nigeria

where the inflation rate is already high at about 15% already, any increase in such rate will lead

to economic problems in the society and a reduction in economic growth.

Correspondingly, interest rate was found to be positively related to economic growth in

the three countries (UK, USA and Nigeria) line with the a-priori expectation which connotes that

an increase in the interest rate at which credit is extended by deposit money banks in the

economy will improve the economy. This is possible because an increase in interest rate will

ensure that credit is only extended to serious borrowers in the economy which will reduce the

rate of non-performing loans in the economy. Also, as expected according to the theoretical

expectation, Money Supply exhibited a positive effect on economic growth in the three countries

(UK, USA and Nigeria). This means that an increase in money supply will spur economic

growth. This is probable because an increase in money supply leans towards the expansionary

monetary policy which will lead to an improvement in financial deepening, innovation and

economic development in all countries as there will be more credit for investment. On the other

120
hand, as expected in conformity with the theoretical expectation, Exchange Rate was found to

exert negative effect on economic growth of USA and Nigeria and a positive effect on economic

growth of UK, this is probable because the rate of exchange as regards Nigeria has led to the

increase in domestic prices and heightened the price of imported goods which Nigeria depends

on so much which has led to a reduction in the patronage of such goods and the rate of economic

activities in the country. Also, as regards the USA, the rate of securing production inputs such as

labour, opening of production centres in other countries and importing vehicles even though they

produce has taken its turn on economic growth. However, the UK has managed its rate of

exchange and has used it to spur economic growth by minimizing imports and participating in

the European Economic Union which makes it easy to trade with other industrialized nations of

the world at a lower price necessitating a growth in the economy.

Furthermore, in line with the specific objective of this study which is to determine the

causal relationship between variables used in the study as well as the direction of causality, the

study employed the Engle Granger Causality Technique for this purpose. Consequently, it was

discovered that economic growth causes Interest Rate while there exists no causality between

each of Monetary Policy Rate, Inflation Rate, Money Supply, Exchange Rate and economic

growth in the Nigerian context which implies that a change in the behaviour of the Nigerian

economy will lead to a change in the behaviour of Interest rate as compared to the USA context

where economic growth causes Monetary Policy Rate and Interest Rate while there exists no

causality between each of Inflation Rate, Money Supply, Exchange Rate and economic growth

which implies that a change in the behaviour of the economy will cause a change in the interest

rate and monetary policy in the USA, hence, it can be implied that monetary policy in the USA is

used as a response to changes within the economy. Also, in UK, it was revealed that there exists

121
a bi-directional causality between economic growth and Monetary Policy Rate and economic

growth causes Interest Rate and Money Supply while there exists no causality between each of

Inflation Rate, Exchange Rate and economic growth in the UK context. This implies that

economic growth and monetary policy causes a change in the behaviour of each other while

economic growth dictates the trend of interest rate and money supply in the UK.

Meanwhile, the test for the statistical significance of the parameters in the long run using

the probability test revealed that monetary policy is not significant in explaining the changes that

may occur in economic growth in UK and USA but has a significant effect on economic growth

in Nigeria through money supply. Also, the LM correlation test, Heteroskedasticity test, stability

test and functionality RESET test implied that there the result is reliable and sufficiently captures

the impact of monetary policy on economic growth in Nigeria, UK and USA.

122
 CHAPTER FIVE

SUMMARY, CONCLUSION AND RECOMMENDATIONS

5.1 Summary of the Research

The objective of this study is to examine the impact of monetary policy and its lag on

economic growth in Nigeria. The study considered various conceptual issues related to the

subject matter, also, various empirical contributions in literature relating to the subject matter

across various countries of the world were reviewed. In addition, the study considered various

theories relating to the subject matter.

The study adopted time series annual data spanning through 1986 to 2016. The data was

culled from secondary sources especially the World Bank database. In the section meant for

analysis, the study employed the Generalized Method of Moments (GMM) technique to

determine the short run and lag effect while the Auto Regressive Distributed Lag (ARDL)

modeling technique was used to examine the long run equilibrium relationship between the

variables as the variables were not found to be co-integrated in the same order. The Augmented

Dickey Fuller Unit Root Test revealed that all variables except inflation and exchange rate which

were stationary at level were stationary at first difference in Nigeria while all variables were

stationary at first difference in case of USA except Money Supply which was stationary at first

difference and all variables were stationary at first difference in the case of UK. As a result of the

mixed order of integration, the ARDL modeling was used.

Evidence from the ARDL Bounds test revealed that there exists a long run stable

relationship between the variables. However, the GMM technique also showed that the lagged

value of Gross Domestic Product has significant effect on economic growth of the three

123
countries; meanwhile, the lagged value of money supply has significant effect on the economy of

UK and USA while money supply has significant effect on economic growth of the USA.

However, inflation and the lagged value of exchange rate were found to exert significant effect

on economic growth of Nigeria. However, it can be deduced that monetary policy rate, its lagged

value and inflation has negative effect on economic growth in the three countries. However,

Money supply and its lagged value had a negative and positive effect respectively on economic

growth in Nigeria while money supply lag had negative effect on economic growth in USA and

UK while money supply has positive effect on economic growth in both countries in the short

run. On the other hand, the ARDL technique in the long run revealed that monetary policy rate,

inflation rate and exchange rate exert negative effect on economic growth in Nigeria as only

monetary policy rate has negative effect on economic growth in the UK while monetary policy

rate and exchange rate has negative effect on economic growth in the USA as other variables

exert positive influence on economic growth in all countries in the long run. However, monetary

policy has no significant effect on economic growth in UK and USA while Money Supply has a

substantial effect on economic growth in Nigeria as a monetary policy variable

Also, the tests carried out to check the robustness and reliability of the model such as the

LM Serial Correlation test, Ramsey Reset Test, CUSUM test and Heteroskedasticity test

revealed that the model has no problem of autocorrelation, heteroskedasticity and instability, as a

result, the findings of the model can be relied upon for policy recommendations and decision

making as the model is statistically significant.

In addition, the study adopted Engle Granger Causality technique to reveal the causal

relationship between the variables revealing that economic growth causes Interest Rate while

there exists no causality between each of Monetary Policy Rate, Inflation Rate, Money Supply,

124
Exchange Rate and economic growth in the Nigerian context as compared to the USA context

where economic growth causes Monetary Policy Rate and Interest Rate while there exists no

causality between each of Inflation Rate, Money Supply, Exchange Rate and economic growth.

Also, in UK, it was revealed that there exists a bi-directional causality between economic growth

and Monetary Policy Rate and economic growth causes Interest Rate and Money Supply while

there exists no causality between each of Inflation Rate, Exchange Rate and economic growth in

the UK context.

5.2 Conclusion of the Research

The study examined the impact of monetary policy on economic growth in Nigeria. The

study used Gross Domestic Product as the dependent variable and also used Monetary policy

rate, inflation rate, interest rate, money supply and exchange rate as independent variables. The

study used the Generalized Methods of Moments and Auto Regressive Distributed Lag modeling

approach for analysis coupled with the Engle Granger causality test to reveal the direction of

causality. The study was carried out form 1986 to 2016.

In summary, it was discovered that monetary policy has no significant effect on economic

growth in the long run in UK and USA but has significant effect on economic growth in Nigeria

through the medium of money supply. However, economic growth causes monetary policy in

USA and Nigeria while both phenomena dictate each other‟s tune in the UK.

5.3 Policy Recommendations

In the light of the findings and conclusion of the study, it is imperative to make proper

policy recommendations to enhance economic growth in Nigeria and abroad. These policies

which can be considered by economic experts and policy makers within the economy include:

125
i. The existence of monetary policy lag has been discovered to exert negative influence

on economic growth in the economy across the world, as a result, all hands must be

on deck by monetary policy makers to reduce the presence of lags in the

implementation of monetary policy while proper public relations should be enhanced

be monetary policy experts to minimize the presence of information lag in the

monetary system.

ii. Furthermore, inflation in Nigeria was found to exert a negative effect on economic

growth due to its high rate contrary to the discovery in the UK and USA. Hence, it is

suggested that monetary experts in Nigeria should develop measures to reduce the

rate of inflation in the economy and maintain such reduction in a bid to establish the

economy and make the same compete favorably with other industrialized economies

of the world.

iii. Also, the study revealed no form of causality between monetary policy rate and

economic growth in Nigeria as compared to USA where economic growth dictates the

trend of monetary policy. In the same vein, it is suggested that the monetary policy

should not just be seen as a mere tool to display the tyranny of monetary authorities in

Nigeria but should be carefully, reasonably and wisely adopted a an effective tool of

response to put the economy on track after a careful observation of the economy has

shown that it is straying away from set targets and goals; until then, the monetary

policy should not be aggressively used as the market force mechanisms should be

allowed to determine prices and phenomena within the economy.

iv. Finally, in Nigeria, money supply was discovered to be the only monetary policy

variable which had significant effect on economic growth. This means that the use of

126
 money supply as a monetary policy instrument and transmission channel will affect

the economy significantly and quickly than other channels like the monetary policy

rate or exchange rate. As a result, monetary authorities are advised to make use of

money supply in sharply correcting the economy in case of any perceived deviation

from set targets as this channel will bring about a substantial effect within the

economy without so much delay.

5.4 Areas of Further Research

Researchers in the field of international and developmental finance carrying out research

inquiring into the subject matter can make use of other methodologies and other variables.

Meanwhile, prospective researchers can also consider the other countries across other continents

like Australia, South America and Asia to assess if the result is sensitive to the countries

considered. Also, daily, weekly or monthly data can be used for analysis to reveal if the results

discovered in the study are responsive to the recurrence of data. However, this research study

will assist other researchers in drawing conclusions about the impact of monetary policy and its

lag on economic growth and the causal relationship between the variables especially from the

standpoint of the three countries considered viz; Nigeria, the United States of America and the

United Kingdom.

127
 REFERENCES

AbuDalu, A., Elsadig, M. A., Almasaled, S. W., & Abuelgasim, I. E. (2014). The Real Effective
Exchnage Rate Impact on ASEAN -5 Economic Growth. International Journal of
Economics and Management Sciences, 3(2), 1-11.
Aftab, A., Mohsin, A. Q., & Saboor, A. (2016). Impact of Monetary Policy in Economic Growth
in Pakistan: Evaluation and Analysis. Pakistan Journal of Social Sciences, 36(1), 131-
140.
Akinjare-Victoria, A. A., Babajide, I. A. A., & Okafor, T. (2016). Monetary Policy and its
Effectiveness on Economic Development in Nigeria. International Business
Management, 10(22), 5336-5340.
Amassoma, D., Nwosa, P. I., & Olaiya, S. A. (2011). An Appraisal of Monetary Policy and Its
Effect on Macro Economic Stabilization in Nigeria. Journal of Emerging Trends in
Economics and Management Sciences, 2(3), 232-237.
Amassoma, J. D., Nwosa, P. I., & Ofere, A. F. (2011). The Nexus of Interest Rate Deregulation,
Lending Rate and Agricultural Productivity in Nigeria. Current Research Journal of
Economic Theory, 3(2), 53-61.
Anowor, O. F., & Okorie, G. C. (2016). A Reassessment of the Impact of Monetary Policy on
Economic Growth: Study of Nigeria. International Journal of Developing and Emerging
Economies, 4(1), 82-90.

Apinran, M. O. (2015). Potency of Monetary Policy Instruments on Economic Growth of

Nigeria. (M.Sc. Thesis). Eastern Mediterranean University. Cyprus.

Arrellano, M., & Bond, S. (1991). Some Tests of Specification for Panel Data: Monte Carlo
Evidence and an Application to Employment Equations. Review of Economic Studies,
58(1), 287-297.

Arrow, K. J. (1962). The Economic Implication of Learning by Doing. Review of Economic

Studies, 29(1), 155-173.
Asghar, N., & Hussain, Z. (2014). The Lags in Effect of Monetary Policy: A Case Study of
Pakistan. Pakistan Economic and Social Review, 52(1), 1-14.

Asteriou, D., & Hall, S. G. (2011). Applied Econometrics. Palgrave Macmillan.

Baghebo, M., & Ebibai, T. S. (2014). Monetary Policy and Economic Growth in Nigeria (1980-
2011). Asian Economic and Financial Review, 4(1), 20-32.

Bank of England Act 1946.

Bazina, A. Y. (2012). The Impact of Monetary Policy on Particular Sectors of the Economy-
Case Study of Turkey and Poland. (Ph.D. Thesis). Czech University of Life Sciences.
Prague.

128
Bokreta, K., & Benanaya, D. (2016). The Fiscal-Monetary Policy and Economic Growth in
Algeria: VECM Approach. International Journal of Social, Educational, Economic,
Business and Industrial Engineering, 10(9), 3074-3078.
Borio, C., & Gambacorta, L. (2017). Monetary Policy and Bank Lending in a Low Interest Rate
Environment: Diminishing Effectiveness? Bank for International Settlements Paper
Series.
Chan, W. (2016). Analysis on Money Supply, Inflation Rate and Economic Growth through
Empirical Study. (B.Sc. Thesis). Hong Kong Baptist University.
Chaudhry, I. S., Qamber, Y., & Farooq, F. (2012). Monetary Policy, Inflation and Economic
Growth in Pakistan: Exploring the Co-Integration and Causality Relationships. Pakistan
Journal of Commerce and Social Sciences, 6(2), 332-347.
Chih-Hsiang, C., Kam, C. & Chan, H. F. (2009). Effect of Money Supply on Real Output and
Price in China. China and World Economy, 17(2), 35-44.

Crowther, G. (1948). An Outline of Money. London: Thomas Nelson and Sons.

Culbertson, J. M. (1960). Friedman on the lag in effect of monetary policy. Journal of Political
Economy, 68(6), 617-621.

Dalhatu, A. (2012). Monetary Policy and Price Stability in Nigeria (December 2006 through
February 2012). (M.Sc. Thesis). Ahamdu Bello University, Zaria.

Dickey, D. A., & Fuller, W. A. (1981). Likelihood Ratio Statistics for Autoregressive Time
Series with a Unit Root. Econometrica, 49(1), 1057-1072.
Douanla, L. T. (2014). Assessing the Effect of Monetary Policy on Economic growth in Franc
Zone. MPRA Paper Series.
Drama, B. G. H. (2017). Estimation of the Impact of Monetary Policy on Economic Growth: The
Case of Cote d‟Ivoire in Line with SVAR Methodology. Applied Economics and
Finance, 4(4), 66-83.

Dritsakis, N. (2011). Demand for Money in Hungary: An ARDL Approach. Review of

Economics and Finance, 5(1), 1-28.

Dwivedi, D. N. (2010). Macroeconomics. Delhi: McGraw Hill Ltd.

Einzig, P. (1964). Monetary Policy: Ends and Means. England: Penguin Books.

Engle, R. F., & Granger, C. W. J. (1987). Co-Integration and Error Correction Representation:
Estimation and Testing. Econometrica, 55(1), 251-276.

Federal Reserve Board Act from Publications 2003.

Fforde, J. (1992). The Role of the Bank of England, 1941-1958.

129
Fiador, V. O. L. (2015). Monetary Policy and Economic Performance-Evidence from Selected
African Countries. (Ph.D. Thesis). University of Cape Town. South Africa.

Fisher, I. (1911). The Purchasing Power of Money. New York: Kelley.

Freedom of Information in Bank of England Publications (2013)

Friedman, M. (1956). The Quantity Theory of Money: A Restatement. In Studies in the Quantity
Theory of Money. Chicago: University of Chicago Press.

Friedman, M. (1961). The lags in the Effect of Monetary Policy. Journal of Political Economy,
69(1-2), 11-18.

Froyen, R. T. (2009). Macroeconomics: Theories and Policies. Upper Saddle River: Pearson
Prentice Hall.

Giles, D. (2013). Econometrics Beat. Dave Giles‟ Blog: ARDL Models-Part 1 and 2.

Granger, C. W. J. (1969). Investigating Causal Relations by Econometric Models and Cross-

Spectral Methods. Econometrica, 37(1), 428-438.

Hair, J. (2010). Multivariate Data Analysis. London: Pearson

Hendry, D. F. (1995). Dynamic Econometrics. Oxford University Press.

Inam, U., & Ime, B. S. (2017). Monetary Policy and Economic Growth in Nigeria: Evidence
from Nigeria. Advances in Social Sciences Research Journal, 4(6), 41-59.
Irfan, H., & Amen, U. (2011). Impact of Monetary Policy on Gross Domestic Product.
Interdisciplinary Journal of Contemporary Research in Business, 3(1), 1348-1361.

Isedu, M. (2013). Effects of Monetary Policy on Macroeconomic Performance: The Case of

Nigeria. (Ph.D. Thesis). University of Greenwich.

Jain, T. R., & Khanna, O. P. (2006). Macroeconomic Management. Delhi: Nee Kunj Print
Process.

Jhingan, M. I. (1997). Macroeconomic Theory. Delhi: Nisha Enterprises.

Johansen, S., & Juselius, K. (1990). Maximum Likelihood Estimation and Inference on Co-
Integration-with Applications to the Demand for Money. Oxford Bulletin of Economics
and Statistics, 52(1), 169-210.

Johnson, H. G. (1978). Selected Essays in Monetary Economics: Monetary Theory and Policy.
American Economic Review, 50(11), 335-384.

Katiricioglu, S. T. (2009). Revisiting the Tourism-Led growth hypothesis for Turkey using the
Bounds Test and Johansen Approach for Co-integration. Tourism Management, 30(1),
17-20.

130
Keynes, J. M. (1936). The General Theory of Employment, Interest and Money. London:
MacMillan Ltd.

Klasco, J. (2013). Mathematical Analysis of the Transmission Mechanism of Monetary Policy of

the National Bank of Slovakia. (Ph.D. thesis). Comenius University. Slovakia.
Kutu, A. A., Nzimande, N. P., & Msomi, S. (2017). Effectiveness of Monetary Policy and the
Growth of Industrial Sector in China. Journal of Economics and Behavioral Studies, 9(3),
46-59.
Lashkary, M., Kashani, B. H. (2011). The Impact of Monetary Variables on Economic Growth in
Iran: A Monetarists‟ Approach. World Applied Sciences Journal, 15(3), 449-456.
Liu, L. (2012). Monetary Policy, Bank Size and Bank Lending: Evidence from Australia. MORA
Paper Series.

Lucas, R. E. (1988). On the Mechanics of Economic Development. Journal of Monetary

Economics, 22(1), 3-42.
McKinnon, R. (1973). Money and Capital in Economic Development. Washington D.C.:
Brookings Institution
Mehdi, S., & Reza, M. (2011). Effects of Monetary Policy on Industry Sector Growth in Iran.
European Journal of Experimental Biology, 1(4), 29-40.

Mishkin, F. (1996). The Channels of Monetary Transmission: Lessons for Monetary Policy.
National Bureau of Economic Research Working Papers.
Mushtaq, S., & Siddiqui, D. A. (2016). Effect of Interest Rate on Bank Deposits: Evidences from
Islamic and Non-Islamic Economies. Future Business Journal, 3(2017), 1-8.
Najaf, R. (2017). IS Monetary Policy Necessary for the Economic Development of Less
Developed Countries with Special Reference to Pakistan. Business and Economics
Journal, 8(1), 1-3.

Narayan, P. K. (2005). The Saving and Investment Nexus for China: Evidence from Co-
Integration Tests. Applied Economics, 37(17), 1979-1990.
Noman, S., & Khudri, M. (2015). The Effects of Monetary and Fiscal Policy on Economic
Growth in Bangladesh.
Obadeyi, J. A., Okhiria, A. O., & Afolabi, V. K. (2016). Evaluating the Impact of Monetary
Policy on the Growth of Emerging Economy: Nigerian Experience. American Journal of
Economics, 6(5), 241-249.
Ogunmuyiwa, M. S., & Ekone, A. F. (2010). Money Supply-Economic Growth Nexus in
Nigeria. Journal of Social Science, 22(3), 199-204.

131
Okwu, A. T., Obiakor, R. T., Falaiye, O. B., & Owolabi, S. A. (2011). Empirical Analysis of the
Effects of Monetary Policy Innovations on Stabilization of Commodity Prices in Nigeria.
European Journal of Economics, Finance and Administrative Sciences, 1(32), 64-79.

Oloyede, J. A. (2002). Principles of International Finance. Forthright Educational Publishers:

Lagos.
Olweny, T., & Chiluwe, M. (2012). The Effect of Monetary Policy on Private Sector Investment
in Kenya. Journal of Applied Finance and Banking, 2(2), 239-287.

Onderi, V. N. (2014). The Impact of Monetary Policy on Economic Growth and Price Stability in
Kenya: 1992-2013. (M.Sc. Thesis). Kenyatta University. Kenya.
Osmond, N. O., Egbulonu, K. G., & Emerenini, F. M. (2015). Monetary Policy and the
Manufacturing Sector in Nigeria. SSRG International Journal of Economics and
Management Sciences, 2(1), 17-25.

Oteng-Abayie, E. F., & Frimpong, J. M. (2006). Bounds Testing Approach to Co-Integration: An

Examination of Foreign Direct Investment, Trade and Growth Relationships. American
Journal of Applied Sciences, 3(11), 2079-2058.

Pesaran, H. M., & Shin, Y. (1999). An Autoregressive Distributed Lag Modeling Approach to
Co-integration Analysis. In S. Strom, A. Holly and P. Diamond (Eds.), Econometrics and
Economic Theory in the 20th Century: The Ranger Frisch Centennial.

Pesaran, M. H., & Pesaran, B. (1997). Working with Microfit 4.0: Interactive Econometric
Analysis. Oxford: Oxford University Press.

Pesaran, M. H., Shin, Y., & Smith R. J. (2001). Bounds Testing Approaches to the Analysis of
Level Relationships. Journal of Applied Econometrics, 16(1), 289-326.
Poon, W. C. (2010). Testing Transmission Mechanisms on Economic Growth in Malaysia.
Monash University Business and Economics Discussion Paper series.

Rebelo, S. (1991). Long Run Policy Analysis and Long Run Growth. Journal of Political
Economy, 99(1), 500-521.
Ridhwan, M. M., Henri, L. F., Nijkamp, P., & Piet, R. (2010). The Impact of Monetary Policy on
Economic Activity-Evidence from a Meta-Analysis. Tinbergen Institute Discussion
Paper Series.

Romer, P. M. (1986). Increasing Returns and Long Run Growth. Journal of Political Economy,
94(5), 1002-1037.

Sanusi, J. O. (2002). The Evolution of Monetary Management in Nigeria and its Impact on
Economic Development. Bullion Publication of the Central Bank of Nigeria.

132
Saqib, N., & Aggarwal, P. (2017). Impact of Fiscal and Monetary Policy on Economic Growth in
an Emerging Economy. International Journal of Applied Business and Economic
Research, 15(4), 457-462.

Schumpeter, J. A. (1911). The Theory of Economic Development, An Inquiry into Profits,

Capital, Credit, Interest, and the Business Cycle. Cambridge MA, Harvard University
Press.

Scitovsky, T. (1940). A Study of Interest and Capital. Economics, 7(27), 293-317.

Shaw, E. S. (1973). Financial Deepening in Economic Development. New York: Oxford

University Press.

Shrestha, M. B. & Chowdhury, K. (2007). Testing Financial Liberalization Hypothesis with

ARDL modeling Approach. Applied Financial Economics, 17(1), 1529-1540.

Shrestha, M. B. (2005). Financial Liberalization in Nepal. Australia: University of Wollongong.

Smitha, T. H. (2010). Impact of Monetary Policy on the Indian Economy in the Post-Reform
Period. (Ph.D. Thesis). Cochin University of Science and Technology. Kerala.
Srithilat, K., & Sun, G. (2017). The Impact of Monetary Policy on Economic Development:
Evidence from Lao PDR. Global Journal of Human-Social Science, 17(2), 9-15.

Sulaiman, L. A. (2006). Monetary and Fiscal Policies Management. Akure: Adeyemo

Publishing House.
Sulaiman, L. A., & Migiro, S. O. (2014). The Nexus between Monetary Policy and Economic
Growth in Nigeria: A Causality Test. Public and Municipal Finance, 3(2), 35-40.
Tang, A. (2009). The Short-Term Impact of Monetary Policy on Economic Growth and Inflation.
(M.Ba. Thesis). University of North Carolina. USA.
ThankGod, O. A., & Karimo, T. M. (2014). Monetary Policy Effectiveness, Output Growth and
Inflation in Nigeria. International Journal of Economics, Finance and Management, 3(6),
301-307.
Wong, K. M., & Chong, T. T. (2015). What Should Central Banks Target? Evidence on the
Impact of Monetary Policy Regimes on Economic Growth. (Unpublished Paper).

133
 APPENDIX I

Table A (1):- THE TABLE DISPLAYING THE RAW DATA SHOWING THE RELATIONSHIP

BETWEEN MONETARY POLICY INDICIES AND ECONOMIC GROWTH IN NIGERIA

EXGR
Year GDP MPR INFL INTR MS 2010=100
1986 1.14E+11 10.00 5.717151 10.00 2.36E+10 267.6441646
1987 1.01E+11 12.75 11.29032 15.80 2.89E+10 85.26640197
1988 1.09E+11 12.75 54.51122 14.30 3.84E+10 85.68371346
1989 1.16E+11 18.50 50.46669 21.20 4.34E+10 76.29959873
1990 1.31E+11 18.50 7.3644 23.00 5.76E+10 70.7944928
1991 1.3E+11 15.50 13.00697 20.10 7.91E+10 60.00861583
1992 1.31E+11 17.50 44.58884 20.50 1.29E+11 49.77730748
1993 1.33E+11 26.00 57.16525 28.02 1.98E+11 54.53863478
1994 1.35E+11 13.50 57.03171 15.00 2.67E+11 100.8618431
1995 1.34E+11 13.50 72.8355 14.27 3.19E+11 160.2339079
1996 1.41E+11 13.50 29.26829 13.55 3.7E+11 207.7710672
1997 1.45E+11 13.50 8.529874 7.43 4.3E+11 236.0781831
1998 1.49E+11 13.50 9.996378 10.09 5.26E+11 272.5217744
1999 1.5E+11 18.00 6.618373 14.30 7E+11 70.19258278
2000 1.57E+11 14.00 6.933292 10.44 1.04E+12 69.91508798
2001 1.64E+11 20.50 18.87365 10.09 1.31E+12 77.88370892
2002 1.71E+11 16.50 12.87658 15.57 1.56E+12 78.12851967
2003 1.88E+11 15.00 14.03178 11.88 1.77E+12 73.24736539
2004 2.52E+11 15.00 14.99803 12.21 2.13E+12 74.95607689
2005 2.61E+11 13.00 17.86349 8.68 2.61E+12 85.54579545
2006 2.82E+11 10.00 8.239527 8.26 3.56E+12 91.50237702
2007 3.01E+11 9.50 5.382224 9.49 5.88E+12 89.64686176
2008 3.2E+11 9.75 11.57798 11.95 9.32E+12 99.1249731
2009 3.42E+11 6.00 11.53767 12.63 1.09E+13 92.13575199
2010 3.69E+11 6.25 13.7202 7.19 1.17E+13 100
2011 3.87E+11 12.00 10.84079 6.30 1.32E+13 100.3082223
2012 4.04E+11 12.00 12.21701 7.63 1.54E+13 111.3890702
2013 4.25E+11 12.00 8.475827 6.72 1.73E+13 118.8129984
2014 4.52E+11 13.00 8.057383 9.89 1.82E+13 127.095076
2015 4.64E+11 11.00 9.017684 8.26 1.86E+13 126.0683542
2016 4.57E+11 14.00 15.69685 5.46 2.06E+13 115.6631368
Source: World Bank Database (2017)

134
 Table A (2):- THE TABLE DISPLAYING THE RAW DATA SHOWING THE RELATIONSHIP

BETWEEN MONETARY POLICY INDICIES AND ECONOMIC GROWTH IN USA

Year GDP MPR INFL INTR MS EXGR 2010=100

1986 7.96E+12 8.3325 1.858736 6.189985 3.53E+12 125.2046029
1987 8.23E+12 8.203333 3.740876 5.511425 3.68E+12 112.6913813
1988 8.58E+12 9.315 4.009088 5.617435 3.93E+12 106.1675311
1989 8.89E+12 10.87333 4.827003 6.723888 4.14E+12 109.6212274
1990 9.06E+12 10.00917 5.397956 6.084919 4.25E+12 104.53538
1991 9.06E+12 8.463333 4.234964 4.969413 4.31E+12 103.1549862
1992 9.38E+12 6.251667 3.02882 3.883588 4.3E+12 100.6915636
1993 9.64E+12 6 2.951657 3.536594 4.33E+12 103.710154
1994 1E+13 7.138333 2.607442 4.905799 4.35E+12 103.1382264
1995 1.03E+13 8.829167 2.80542 6.605716 4.65E+12 99.66617492
1996 1.07E+13 8.270833 2.931204 6.329727 5.01E+12 102.6537135
1997 1.12E+13 8.441667 2.33769 6.616914 5.41E+12 107.6706423
1998 1.17E+13 8.354167 1.552279 7.190871 5.93E+12 115.3227923
1999 1.22E+13 7.994167 2.188027 6.366418 6.5E+12 114.2681339
2000 1.27E+13 9.233333 3.376857 6.80301 7.02E+12 117.945007
2001 1.28E+13 6.921667 2.826171 4.539319 7.58E+12 124.6046275
2002 1.31E+13 4.675 1.586032 3.092403 7.92E+12 124.296867
2003 1.34E+13 4.1225 2.270095 2.086831 8.27E+12 116.3795002
2004 1.39E+13 4.34 2.677237 1.547721 8.74E+12 110.9257116
2005 1.44E+13 6.189167 3.392747 2.878896 9.45E+12 109.3673132
2006 1.48E+13 7.9575 3.225944 4.73962 1.03E+13 108.7484141
2007 1.51E+13 8.05 2.852672 5.248971 1.15E+13 103.6216645
2008 1.5E+13 5.0875 3.8391 3.065749 1.24E+13 99.54347993
2009 1.46E+13 3.25 -0.35555 2.471794 1.31E+13 104.0383525
2010 1.5E+13 3.25 1.640043 2.004173 1.28E+13 100
2011 1.52E+13 3.25 3.156842 1.161394 1.36E+13 95.09801322
2012 1.55E+13 3.25 2.069337 1.382482 1.43E+13 97.99889924
2013 1.58E+13 3.25 1.464833 1.609007 1.49E+13 99.11466237
2014 1.62E+13 3.25 1.622223 1.433975 1.57E+13 101.1806041
2015 1.66E+13 3.26 0.118627 2.160831 1.62E+13 113.8385707
2016 1.69E+13 3.511667 1.261583 2.168002 1.68E+13 117.5139307
Source: World Bank Database (2017)

135
 Table A (3):- THE TABLE DISPLAYING THE RAW DATA SHOWING THE RELATIONSHIP

BETWEEN MONETARY POLICY INDICIES AND ECONOMIC GROWTH IN UK

Year GDP MPR INFL INTR MS EXGR 2010=100

1986 1.42E+12 10.90136 6.296171 1.78E+11 95.4838566
1987 1.5E+12 9.7355 4.040502 3.38E+11 95.35635034
1988 1.59E+12 10.09571 3.787497 3.94E+11 102.9495308
1989 1.63E+12 13.84181 5.237595 5.47716 4.73E+11 103.3386058
1990 1.64E+12 14.76812 6.972683 6.361031 5.23E+11 106.8567641
1991 1.62E+12 11.69498 7.532649 4.940826 5.32E+11 108.8736485
1992 1.63E+12 9.563483 4.261548 5.99833 3.8E+11 105.014797
1993 1.67E+12 6.013633 2.5065 3.210615 3.97E+11 94.05871639
1994 1.73E+12 5.462783 1.97849 4.052359 4.28E+11 94.24131362
1995 1.78E+12 6.693675 2.656452 -2.58669 5.15E+11 90.84750764
1996 1.82E+12 5.958533 2.481101 1.813451 5.89E+11 92.56700594
1997 1.88E+12 6.5634 1.777946 4.516655 7.39E+11 108.2356526
1998 1.94E+12 7.23205 1.588924 5.974525 8.94E+11 115.189543
1999 2E+12 5.344408 1.335407 4.445903 9.33E+11 114.8791311
2000 2.08E+12 5.963892 0.785269 3.864516 1.04E+12 118.0071583
2001 2.13E+12 5.122375 1.235895 4.161138 1.13E+12 115.4842659
2002 2.18E+12 4 1.256192 1.763821 1.18E+12 116.4559885
2003 2.26E+12 3.695792 1.362922 1.262903 1.3E+12 113.3721404
2004 2.32E+12 4.379817 1.344596 1.898664 1.43E+12 119.9655142
2005 2.39E+12 4.648675 2.049668 1.93274 1.63E+12 119.5165722
2006 2.45E+12 4.638258 2.333528 1.643616 1.85E+12 121.5019002
2007 2.51E+12 5.508708 2.321036 2.886834 2.15E+12 125.7127081
2008 2.49E+12 4.675308 3.613499 1.800291 2.53E+12 110.1397089
2009 2.38E+12 0.644867 2.166231 -0.87221 2.53E+12 96.54687864
2010 2.43E+12 0.5 3.285714 -1.02625 2.63E+12 100
2011 2.47E+12 0.5 4.48424 -1.48164 2.51E+12 101.4781971
2012 2.5E+12 0.5 2.82171 -1.02339 2.53E+12 106.8318691
2013 2.55E+12 0.5 2.554547 -1.37807 2.59E+12 105.7679573
2014 2.62E+12 0.5 1.460192 -1.12699 2.52E+12 113.7006279
2015 2.68E+12 0.050021 2.57E+12 121.7959757
2016 2.73E+12 0.641613 2.79E+12 109.721221
Source: World Bank Database (2017)

136
Table A (4):- THE TABLE DISPLAYING THE LOGLINEARIZED RAW DATA SHOWING THE

RELATIONSHIP BETWEEN MONETARY POLICY INDICIES AND ECONOMIC

GROWTH IN NIGERIA

EXGR
Year GDP MPR INFL INTR MS 2010=100
1986 25.45624 2.302585 1.743471 2.302585 23.88473 5.589658
1987 25.3425 2.545531 2.423946 2.76001 24.08695 4.44578
1988 25.41521 2.545531 3.998407 2.66026 24.37147 4.450663
1989 25.47788 2.917771 3.921313 3.054001 24.49305 4.334668
1990 25.59802 2.917771 1.996658 3.135494 24.77598 4.259781
1991 25.59183 2.74084 2.565486 3.00072 25.09357 4.094488
1992 25.59615 2.862201 3.797484 3.020425 25.58374 3.907559
1993 25.61684 3.258097 4.045946 3.332919 26.01395 3.998909
1994 25.6259 2.60269 4.043607 2.70805 26.31031 4.613752
1995 25.62282 2.60269 4.288204 2.658159 26.48772 5.076635
1996 25.67155 2.60269 3.376505 2.606202 26.63767 5.336437
1997 25.69919 2.60269 2.143575 2.005189 26.78643 5.464163
1998 25.72598 2.60269 2.302223 2.311793 26.98788 5.607719
1999 25.73071 2.890372 1.88985 2.66026 27.27397 4.251243
2000 25.78253 2.639057 1.936335 2.345645 27.66647 4.247281
2001 25.82569 3.020425 2.937767 2.311793 27.90056 4.355217
2002 25.86284 2.80336 2.55541 2.745346 28.07301 4.358355
2003 25.96137 2.70805 2.641325 2.474856 28.19974 4.293842
2004 26.25206 2.70805 2.707919 2.502255 28.38769 4.316902
2005 26.28593 2.564949 2.882759 2.161022 28.59148 4.449052
2006 26.36484 2.302585 2.108943 2.111425 28.90154 4.516365
2007 26.43089 2.251292 1.683102 2.249712 29.40186 4.495878
2008 26.49171 2.277267 2.449105 2.480941 29.86266 4.596381
2009 26.55876 1.791759 2.445618 2.535943 30.02149 4.523263
2010 26.63423 1.832581 2.618869 1.97301 30.0872 4.60517
2011 26.68195 2.484907 2.383316 1.840996 30.20935 4.608248
2012 26.72385 2.484907 2.502829 2.031468 30.36461 4.713029
2013 26.77639 2.484907 2.137218 1.904375 30.48186 4.777551
2014 26.83758 2.564949 2.086589 2.291524 30.53079 4.844935
2015 26.86376 2.397895 2.199188 2.111937 30.55396 4.836824
2016 26.84823 2.639057 2.75346 1.697449 30.65747 4.750682
Source: Microsoft Excel 2010 Output

137
Table A (5):- THE TABLE DISPLAYING THE LOGLINEARIZED RAW DATA SHOWING THE

RELATIONSHIP BETWEEN MONETARY POLICY INDICIES AND ECONOMIC

GROWTH IN USA

EXGR
Year GDP MPR INFL INTR MS 2010=100
1986 29.70501 2.120164 0.619897 1.822933 28.89349 4.829949
1987 29.73904 2.104541 1.31932 1.706823 28.93333 4.724653
1988 29.78022 2.231626 1.388564 1.725875 28.99904 4.665018
1989 29.81637 2.386313 1.574226 1.905667 29.05084 4.697031
1990 29.83538 2.303501 1.68602 1.805813 29.07727 4.649526
1991 29.83464 2.135743 1.443375 1.603302 29.09217 4.636233
1992 29.86957 1.832848 1.108173 1.356759 29.09063 4.612062
1993 29.89666 1.791759 1.082367 1.263164 29.09658 4.6416
1994 29.93624 1.965479 0.95837 1.590418 29.10062 4.63607
1995 29.96307 2.178061 1.031553 1.887935 29.16705 4.601826
1996 30.00033 2.112735 1.075413 1.845257 29.24277 4.631361
1997 30.04422 2.13318 0.849163 1.889629 29.31976 4.679077
1998 30.08776 2.12276 0.439724 1.972812 29.41145 4.747735
1999 30.13354 2.078712 0.783 1.851037 29.50228 4.738548
2000 30.17365 2.22282 1.216945 1.917365 29.58026 4.770218
2001 30.18336 1.934657 1.038923 1.512777 29.6566 4.825146
2002 30.20107 1.542229 0.461235 1.128949 29.70016 4.822673
2003 30.22875 1.41646 0.819822 0.735647 29.74325 4.756856
2004 30.26591 1.467874 0.984785 0.436783 29.79909 4.708861
2005 30.29881 1.8228 1.22164 1.057407 29.87673 4.694712
2006 30.32513 2.074115 1.171226 1.555957 29.96293 4.689037
2007 30.34276 2.085672 1.048256 1.658032 30.0737 4.640746
2008 30.33984 1.626787 1.345238 1.120292 30.15228 4.600595
2009 30.31169 1.178655 #NUM! 0.904944 30.2059 4.64476
2010 30.33669 1.178655 0.494723 0.695231 30.1781 4.60517
2011 30.35258 1.178655 1.149572 0.149621 30.24271 4.554908
2012 30.37458 1.178655 0.727228 0.323881 30.29056 4.584956
2013 30.39121 1.178655 0.381741 0.475617 30.33327 4.596277
2014 30.41464 1.178655 0.483797 0.36045 30.38284 4.616907
2015 30.44027 1.181727 -2.13177 0.770493 30.41636 4.734781
2016 30.4563 1.256091 0.232367 0.773806 30.45327 4.766557
Source: Microsoft Excel 2010 Output

138
Table A (6):- THE TABLE DISPLAYING THE LOGLINEARIZED RAW DATA SHOWING THE

RELATIONSHIP BETWEEN MONETARY POLICY INDICIES AND ECONOMIC

GROWTH IN UK

EXGR
Year GDP MPR INFL INTR MS 2010=100
1986 27.98395 2.388887 #NUM! 1.839942 25.90706 4.558957
1987 28.03615 2.275779 #NUM! 1.396369 26.54539 4.557621
1988 28.09242 2.31211 #NUM! 1.331705 26.69959 4.634239
1989 28.11792 2.627694 1.655862 1.700587 26.88296 4.638011
1990 28.12506 2.692471 1.942 1.85019 26.98299 4.671489
1991 28.11381 2.45916 2.019247 1.597532 26.99947 4.690188
1992 28.1174 2.257952 1.449633 1.791481 26.66218 4.654101
1993 28.14216 1.794029 0.918887 1.166463 26.70807 4.543919
1994 28.18028 1.697958 0.682334 1.399299 26.78183 4.545859
1995 28.20503 1.901163 0.976991 #NUM! 26.96657 4.509182
1996 28.2302 1.784824 0.908702 0.595232 27.10203 4.527933
1997 28.26099 1.881509 0.575459 1.507772 27.32921 4.684311
1998 28.2924 1.978523 0.463057 1.787505 27.51855 4.746579
1999 28.32471 1.676051 0.289236 1.491983 27.56158 4.743881
2000 28.36147 1.785723 -0.24173 1.351836 27.66651 4.770745
2001 28.38837 1.633618 0.211795 1.425789 27.75108 4.749134
2002 28.41206 1.386294 0.228085 0.567483 27.79981 4.757513
2003 28.44613 1.307195 0.309631 0.233413 27.89328 4.730676
2004 28.4711 1.477007 0.296093 0.64115 27.98697 4.787204
2005 28.50038 1.536582 0.717678 0.658938 28.11668 4.783455
2006 28.52511 1.534339 0.847381 0.496899 28.24856 4.79993
2007 28.55034 1.70633 0.842014 1.06016 28.39535 4.833999
2008 28.54405 1.542295 1.284677 0.587948 28.5592 4.70175
2009 28.49981 -0.43871 0.772989 #NUM! 28.55855 4.570029
2010 28.51878 -0.69315 1.189584 #NUM! 28.59768 4.60517
2011 28.53376 -0.69315 1.500569 #NUM! 28.55249 4.619844
2012 28.5468 -0.69315 1.037343 #NUM! 28.56016 4.671256
2013 28.56573 -0.69315 0.937875 #NUM! 28.58096 4.661248
2014 28.59598 -0.69315 0.378568 #NUM! 28.55529 4.733569
2015 28.61768 #NUM! -2.99532 #NUM! 28.57484 4.802347
2016 28.63558 #NUM! -0.44377 #NUM! 28.65842 4.697943
Source: Microsoft Excel 2010 Output

139
 APPENDIX II (NIGERIA)

Table B:- GENERALIZED METHOD OF MOMENTS (GMM) RESULT

Dependent Variable: GDP

Method: Generalized Method of Moments
Date: 09/23/17 Time: 08:37
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments
Linear estimation with 1 weight update
Estimation weighting matrix: HAC (Bartlett kernel, Newey-West fixed
bandwidth = 4.0000)
Standard errors & covariance computed using estimation weighting matrix
Instrument specification: GDP(-1) MPR MPR(-1) INFL INFL(-1) INTR INTR(
-1) MS MS(-1) EXGR EXGR(-1)
Constant added to instrument list

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.844037 0.087529 9.642913 0.0000

MPR -0.011534 0.023149 -0.498225 0.6244
MPR(-1) -0.050199 0.037712 -1.331101 0.1998
INFL -0.042981 0.012816 -3.353727 0.0035
INFL(-1) 0.016762 0.011385 1.472237 0.1582
INTR 0.032350 0.022029 1.468488 0.1592
INTR(-1) -0.033883 0.040483 -0.836984 0.4136
MS -0.118146 0.074160 -1.593129 0.1285
MS(-1) 0.147576 0.085949 1.717024 0.1031
EXGR 0.016802 0.022311 0.753093 0.4611
EXGR(-1) -0.113942 0.044501 -2.560424 0.0197
C 4.001395 2.194582 1.823306 0.0849

R-squared 0.991638 Mean dependent var 26.06324

Adjusted R-squared 0.986527 S.D. dependent var 0.507542
S.E. of regression 0.058911 Sum squared resid 0.062470
Durbin-Watson stat 1.911456 J-statistic 2.30E-36
Instrument rank 12

140
Table C:- RESIDUAL SERIES/ECM

Last updated:
09/22/17 - 12:40
Modified: 1986
2016 //
makeresids ecm

1986 NA
1987 0.115831
1988 0.112779
1989 0.212602
1990 0.275192
1991 0.149365
1992 0.035208
1993 0.041826
1994 -0.068679
1995 -0.062304
1996 -0.020582
1997 -0.081590
1998 -0.044621
1999 -0.196108
2000 -0.288180
2001 -0.258907
2002 -0.222571
2003 -0.199721
2004 0.056058
2005 0.008143
2006 0.002090
2007 -0.029612
2008 -0.028117
2009 -0.035593
2010 -0.030344
2011 0.026624
2012 0.070104
2013 0.089812
2014 0.201768
2015 0.186827
2016 0.106295

141
Table D:- ADF UNIT ROOT TEST (URT) RESULTS

Table D (1):- ADF URT FOR GDP @ LEVEL (NON-STATIONARY)

Null Hypothesis: GDP has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic 0.586226 0.9869

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 12:47
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.013877 0.023672 0.586226 0.5624

C -0.314636 0.615972 -0.510795 0.6135

R-squared 0.012125 Mean dependent var 0.046399

Adjusted R-squared -0.023156 S.D. dependent var 0.062612
S.E. of regression 0.063332 Akaike info criterion -2.616496
Sum squared resid 0.112308 Schwarz criterion -2.523083
Log likelihood 41.24744 Hannan-Quinn criter. -2.586612
F-statistic 0.343661 Durbin-Watson stat 1.569875
Prob(F-statistic) 0.562420

Table D (2):- ADF URT FOR GDP @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(GDP) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.243327 0.0002

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

142
*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP,2)
Method: Least Squares
Date: 09/22/17 Time: 12:47
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(GDP(-1)) -0.892828 0.170279 -5.243327 0.0000

C 0.046720 0.013342 3.501812 0.0016

R-squared 0.504519 Mean dependent var 0.003387

Adjusted R-squared 0.486168 S.D. dependent var 0.078685
S.E. of regression 0.056403 Akaike info criterion -2.846120
Sum squared resid 0.085895 Schwarz criterion -2.751824
Log likelihood 43.26874 Hannan-Quinn criter. -2.816587
F-statistic 27.49247 Durbin-Watson stat 1.797223
Prob(F-statistic) 0.000016

Table D (3):- ADF URT FOR MPR @ LEVEL (NON-STATIONARY)

Null Hypothesis: MPR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.615724 0.1010

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR)
Method: Least Squares
Date: 09/22/17 Time: 12:47
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) -0.378651 0.144760 -2.615724 0.0142

C 0.987013 0.375634 2.627593 0.0138

143
R-squared 0.196372 Mean dependent var 0.011216
Adjusted R-squared 0.167671 S.D. dependent var 0.264034
S.E. of regression 0.240883 Akaike info criterion 0.055333
Sum squared resid 1.624695 Schwarz criterion 0.148746
Log likelihood 1.170000 Hannan-Quinn criter. 0.085217
F-statistic 6.842012 Durbin-Watson stat 2.039573
Prob(F-statistic) 0.014185

Table D (4):- ADF URT FOR MPR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(MPR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -6.642865 0.0000

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR,2)
Method: Least Squares
Date: 09/22/17 Time: 12:48
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MPR(-1)) -1.240585 0.186755 -6.642865 0.0000

C 0.004016 0.048642 0.082557 0.9348

R-squared 0.620401 Mean dependent var -6.15E-05

Adjusted R-squared 0.606342 S.D. dependent var 0.417459
S.E. of regression 0.261923 Akaike info criterion 0.224940
Sum squared resid 1.852299 Schwarz criterion 0.319236
Log likelihood -1.261627 Hannan-Quinn criter. 0.254472
F-statistic 44.12765 Durbin-Watson stat 2.084892
Prob(F-statistic) 0.000000

Table D (5):- ADF URT FOR INFL @ LEVEL (STATIONARY)

Null Hypothesis: INFL has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

144
 t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.966876 0.0497

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL)
Method: Least Squares
Date: 09/22/17 Time: 12:48
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INFL(-1) -0.448937 0.151316 -2.966876 0.0061

C 1.242997 0.423141 2.937551 0.0066

R-squared 0.239179 Mean dependent var 0.033666

Adjusted R-squared 0.212007 S.D. dependent var 0.700840
S.E. of regression 0.622129 Akaike info criterion 1.953000
Sum squared resid 10.83723 Schwarz criterion 2.046414
Log likelihood -27.29501 Hannan-Quinn criter. 1.982884
F-statistic 8.802355 Durbin-Watson stat 1.553814
Prob(F-statistic) 0.006096

Table D (6):- ADF URT FOR INTR @ LEVEL (NON-STATIONARY)

Null Hypothesis: INTR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.762994 0.3908

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR)
Method: Least Squares
Date: 09/22/17 Time: 12:49
Sample (adjusted): 1987 2016

145
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INTR(-1) -0.251305 0.142544 -1.762994 0.0888

C 0.602130 0.357136 1.685996 0.1029

R-squared 0.099914 Mean dependent var -0.020171

Adjusted R-squared 0.067768 S.D. dependent var 0.308186
S.E. of regression 0.297560 Akaike info criterion 0.477939
Sum squared resid 2.479176 Schwarz criterion 0.571352
Log likelihood -5.169090 Hannan-Quinn criter. 0.507823
F-statistic 3.108147 Durbin-Watson stat 1.926989
Prob(F-statistic) 0.088814

Table D (7):- ADF URT FOR INTR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(INTR) has a unit root

Exogenous: Constant
Lag Length: 3 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.774111 0.0086

Test critical values: 1% level -3.711457
5% level -2.981038
10% level -2.629906

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR,2)
Method: Least Squares
Date: 09/22/17 Time: 12:49
Sample (adjusted): 1991 2016
Included observations: 26 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INTR(-1)) -1.955428 0.518116 -3.774111 0.0011

D(INTR(-1),2) 0.696275 0.428002 1.626803 0.1187
D(INTR(-2),2) 0.256063 0.314609 0.813908 0.4248
D(INTR(-3),2) 0.275411 0.192999 1.427011 0.1683
C -0.073178 0.056605 -1.292775 0.2101

R-squared 0.700784 Mean dependent var -0.019076

Adjusted R-squared 0.643790 S.D. dependent var 0.469623
S.E. of regression 0.280287 Akaike info criterion 0.465034
Sum squared resid 1.649773 Schwarz criterion 0.706975
Log likelihood -1.045438 Hannan-Quinn criter. 0.534704
F-statistic 12.29583 Durbin-Watson stat 2.088822
Prob(F-statistic) 0.000026

146
Table D (8):- ADF URT FOR MS @ LEVEL (NON-STATIONARY)

Null Hypothesis: MS has a unit root

Exogenous: Constant
Lag Length: 2 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.646224 0.4466

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS)
Method: Least Squares
Date: 09/22/17 Time: 12:49
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MS(-1) -0.015795 0.009595 -1.646224 0.1128

D(MS(-1)) 0.824183 0.181015 4.553124 0.0001
D(MS(-2)) -0.437558 0.182563 -2.396759 0.0247
C 0.578041 0.282156 2.048657 0.0516

R-squared 0.564149 Mean dependent var 0.224500

Adjusted R-squared 0.509668 S.D. dependent var 0.133674
S.E. of regression 0.093603 Akaike info criterion -1.767941
Sum squared resid 0.210277 Schwarz criterion -1.577626
Log likelihood 28.75117 Hannan-Quinn criter. -1.709760
F-statistic 10.35490 Durbin-Watson stat 1.932968
Prob(F-statistic) 0.000146

147
Table D (9):- ADF URT FOR MS @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(MS) has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.284855 0.0254

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS,2)
Method: Least Squares
Date: 09/22/17 Time: 12:49
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MS(-1)) -0.528478 0.160883 -3.284855 0.0030

D(MS(-1),2) 0.435333 0.188698 2.307039 0.0296
C 0.118379 0.041948 2.822057 0.0092

R-squared 0.314700 Mean dependent var -0.006465

Adjusted R-squared 0.259876 S.D. dependent var 0.112462
S.E. of regression 0.096752 Akaike info criterion -1.732383
Sum squared resid 0.234022 Schwarz criterion -1.589647
Log likelihood 27.25337 Hannan-Quinn criter. -1.688748
F-statistic 5.740183 Durbin-Watson stat 1.904658
Prob(F-statistic) 0.008882

Table D (10):-ADF URT FOR EXGR @ LEVEL (STATIONARY)

Null Hypothesis: EXGR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

148
Augmented Dickey-Fuller test statistic -3.281138 0.0249
Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXGR)
Method: Least Squares
Date: 09/22/17 Time: 12:50
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

EXGR(-1) -0.445951 0.135913 -3.281138 0.0028

C 2.022959 0.627810 3.222247 0.0032

R-squared 0.277715 Mean dependent var -0.027966

Adjusted R-squared 0.251919 S.D. dependent var 0.371370
S.E. of regression 0.321204 Akaike info criterion 0.630858
Sum squared resid 2.888811 Schwarz criterion 0.724271
Log likelihood -7.462873 Hannan-Quinn criter. 0.660742
F-statistic 10.76587 Durbin-Watson stat 1.229386
Prob(F-statistic) 0.002771

Table D (11):-ADF URT FOR ECM @ LEVEL (NON-STATIONARY)

Null Hypothesis: ECM has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.592624 0.4735

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM)
Method: Least Squares
Date: 09/22/17 Time: 12:50
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

149
 ECM(-1) -0.169860 0.106654 -1.592624 0.1229
C -0.000228 0.014879 -0.015291 0.9879

R-squared 0.085875 Mean dependent var -0.000329

Adjusted R-squared 0.052019 S.D. dependent var 0.082292
S.E. of regression 0.080123 Akaike info criterion -2.144033
Sum squared resid 0.173332 Schwarz criterion -2.049737
Log likelihood 33.08848 Hannan-Quinn criter. -2.114500
F-statistic 2.536451 Durbin-Watson stat 1.701366
Prob(F-statistic) 0.122886

Table D (12):-ADF URT FOR ECM @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(ECM) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.674211 0.0009

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM,2)
Method: Least Squares
Date: 09/22/17 Time: 12:51
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(ECM(-1)) -0.931069 0.199193 -4.674211 0.0001

C -0.000406 0.016109 -0.025225 0.9801

R-squared 0.456615 Mean dependent var -0.002767

Adjusted R-squared 0.435716 S.D. dependent var 0.113421
S.E. of regression 0.085201 Akaike info criterion -2.018863
Sum squared resid 0.188739 Schwarz criterion -1.923705
Log likelihood 30.26408 Hannan-Quinn criter. -1.989772
F-statistic 21.84825 Durbin-Watson stat 1.904257
Prob(F-statistic) 0.000079

150
Table E (1):- ARDL BOUNDS TEST

ARDL Bounds Test

Date: 09/22/17 Time: 12:52
Sample: 1988 2016
Included observations: 28
Null Hypothesis: No long-run relationships exist

Test Statistic Value k

F-statistic 4.520553 6

Critical Value Bounds

Significance I0 Bound I1 Bound

10% 1.75 2.87

5% 2.04 3.24
2.5% 2.32 3.59
1% 2.66 4.05

Test Equation:
Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 12:52
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) -0.035986 0.045982 -0.782608 0.4426

INFL(-1) 0.004707 0.017374 0.270915 0.7891
INTR(-1) -0.017362 0.050781 -0.341896 0.7358
MS(-1) -0.009173 0.015290 -0.599947 0.5550
EXGR(-1) -0.044475 0.030868 -1.440801 0.1644
ECM(-2) -0.206648 0.087663 -2.357294 0.0282
GDP(-1) 0.024246 0.024512 0.989136 0.3339

R-squared 0.263682 Mean dependent var 0.051179

Adjusted R-squared 0.053305 S.D. dependent var 0.056669
S.E. of regression 0.055138 Akaike info criterion -2.745639
Sum squared resid 0.063844 Schwarz criterion -2.412588
Log likelihood 45.43895 Hannan-Quinn criter. -2.643822
Durbin-Watson stat 2.205564

151
Table E (2):- LONG RUN RESULT

ARDL Cointegrating And Long Run Form

Dependent Variable: GDP
Selected Model: ARDL(1, 0, 0, 0, 0, 0)
Date: 10/08/17 Time: 14:24
Sample: 1986 2016
Included observations: 30

Cointegrating Form

Variable Coefficient Std. Error t-Statistic Prob.

D(MPR) -0.018199 0.048583 -0.374594 0.7114

D(INFL) -0.003083 0.018438 -0.167190 0.8687
D(INTR) 0.026318 0.055002 0.478498 0.6368
D(MS) 0.041291 0.018706 2.207408 0.0375
D(EXGR) -0.024178 0.036404 -0.664165 0.5132
CointEq(-1) -0.141960 0.078817 -1.801135 0.0848

Cointeq = GDP - (-0.1282*MPR -0.0217*INFL + 0.1854*INTR + 0.2909*MS

-0.1703*EXGR + 18.9642 )

Long Run Coefficients

Variable Coefficient Std. Error t-Statistic Prob.

MPR -0.128198 0.333066 -0.384904 0.7038

INFL -0.021715 0.130889 -0.165903 0.8697
INTR 0.185392 0.413274 0.448594 0.6579
MS 0.290865 0.083859 3.468484 0.0021
EXGR -0.170319 0.257754 -0.660779 0.5153
C 18.964158 3.767020 5.034260 0.0000

Table F:- LM SERIAL CORRELATION TEST

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.338172 Prob. F(2,20) 0.7171

Obs*R-squared 0.948605 Prob. Chi-Square(2) 0.6223

Test Equation:
Dependent Variable: RESID
Method: ARDL
Date: 09/22/17 Time: 12:52
Sample: 1988 2016
Included observations: 29

152
Presample missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) -0.001236 0.027802 -0.044446 0.9650

MPR -0.002163 0.046890 -0.046124 0.9637
INFL 0.000549 0.018233 0.030117 0.9763
INTR 0.005440 0.051756 0.105110 0.9173
MS 0.000752 0.016886 0.044542 0.9649
EXGR 0.000408 0.033475 0.012175 0.9904
ECM(-1) -0.025118 0.102719 -0.244530 0.8093
RESID(-1) 0.199033 0.258781 0.769117 0.4508
RESID(-2) -0.047078 0.269793 -0.174497 0.8632

R-squared 0.032711 Mean dependent var -3.48E-05

Adjusted R-squared -0.354205 S.D. dependent var 0.049693
S.E. of regression 0.057828 Akaike info criterion -2.613572
Sum squared resid 0.066881 Schwarz criterion -2.189239
Log likelihood 46.89680 Hannan-Quinn criter. -2.480676
Durbin-Watson stat 1.997677

Table G: - HETEROSKEDASTICITY TEST

Heteroskedasticity Test: Breusch-Pagan-Godfrey

F-statistic 0.435617 Prob. F(7,21) 0.8687

Obs*R-squared 3.677035 Prob. Chi-Square(7) 0.8161
Scaled explained SS 11.50814 Prob. Chi-Square(7) 0.1179

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 09/22/17 Time: 12:53
Sample: 1988 2016
Included observations: 29

Variable Coefficient Std. Error t-Statistic Prob.

C -0.252911 0.813170 -0.311019 0.7589

GDP(-1) 0.013252 0.041518 0.319192 0.7527
MPR 0.002248 0.006811 0.330102 0.7446
INFL -0.001263 0.002835 -0.445646 0.6604
INTR 0.002313 0.007889 0.293249 0.7722
MS -0.003074 0.010130 -0.303403 0.7646
EXGR -0.002584 0.005181 -0.498777 0.6231
ECM(-1) -0.028278 0.042739 -0.661631 0.5154

R-squared 0.126794 Mean dependent var 0.002384

Adjusted R-squared -0.164274 S.D. dependent var 0.008002
S.E. of regression 0.008635 Akaike info criterion -6.437143
Sum squared resid 0.001566 Schwarz criterion -6.059958
Log likelihood 101.3386 Hannan-Quinn criter. -6.319013
F-statistic 0.435617 Durbin-Watson stat 2.188826

153
Prob(F-statistic) 0.868693

Table H: - RAMSEY RESET TEST

Ramsey RESET Test

Equation: UNTITLED
Specification: GDP GDP(-1) MPR INFL INTR MS EXGR ECM(-1)
Omitted Variables: Squares of fitted values

Value df Probability
t-statistic 1.753287 21 0.0941
F-statistic 3.074015 (1, 21) 0.0941

F-test summary:
Mean
Sum of Sq. df Squares
Test SSR 0.008829 1 0.008829
Restricted SSR 0.069143 22 0.003143
Unrestricted SSR 0.060314 21 0.002872

Unrestricted Test Equation:

Dependent Variable: GDP
Method: ARDL
Date: 09/22/17 Time: 12:53
Sample: 1988 2016
Included observations: 29
Maximum dependent lags: 1 (Automatic selection)
Model selection method: Akaike info criterion (AIC)
Dynamic regressors (0 lag, automatic):
Fixed regressors:

Variable Coefficient Std. Error t-Statistic Prob.*

GDP(-1) 0.788555 0.135286 5.828823 0.0000

MPR -0.011069 0.045829 -0.241521 0.8115
INFL -0.017361 0.016401 -1.058538 0.3018
INTR 0.031176 0.047772 0.652596 0.5211
MS -0.106699 0.057709 -1.848906 0.0786
EXGR 0.006269 0.036892 0.169919 0.8667
ECM(-1) -0.494169 0.207113 -2.385989 0.0265
FITTED^2 0.012502 0.007130 1.753287 0.0941

R-squared 0.991300 Mean dependent var 26.08809

Adjusted R-squared 0.988401 S.D. dependent var 0.497601
S.E. of regression 0.053592 Akaike info criterion -2.785890
Sum squared resid 0.060314 Schwarz criterion -2.408705
Log likelihood 48.39541 Hannan-Quinn criter. -2.667761
Durbin-Watson stat 1.778502

*Note: p-values and any subsequent tests do not account for model
selection.

154
Table I:- CUSUM TEST

15

10

-5

-10

-15
94 96 98 00 02 04 06 08 10 12 14 16

CUSUM 5% Significance

Table J:- ENGLE GRANGER CAUSALITY TEST

Pairwise Granger Causality Tests

Date: 09/22/17 Time: 12:54
Sample: 1986 2016
Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

MPR does not Granger Cause GDP 29 0.15150 0.8602

GDP does not Granger Cause MPR 1.51394 0.2403

INFL does not Granger Cause GDP 29 0.19414 0.8248

GDP does not Granger Cause INFL 2.28891 0.1231

INTR does not Granger Cause GDP 29 0.25095 0.7801

GDP does not Granger Cause INTR 3.62778 0.0420

MS does not Granger Cause GDP 29 1.32348 0.2849

GDP does not Granger Cause MS 0.48099 0.6240

EXGR does not Granger Cause GDP 29 0.61308 0.5499

GDP does not Granger Cause EXGR 0.46142 0.6359

155
INFL does not Granger Cause MPR 29 0.71074 0.5013
MPR does not Granger Cause INFL 2.05364 0.1502

INTR does not Granger Cause MPR 29 1.47618 0.2485

MPR does not Granger Cause INTR 1.29352 0.2927

MS does not Granger Cause MPR 29 2.39290 0.1128

MPR does not Granger Cause MS 0.07086 0.9318

EXGR does not Granger Cause MPR 29 0.03451 0.9661

MPR does not Granger Cause EXGR 0.05370 0.9478

INTR does not Granger Cause INFL 29 10.1738 0.0006

INFL does not Granger Cause INTR 1.33263 0.2826

MS does not Granger Cause INFL 29 3.87569 0.0348

INFL does not Granger Cause MS 3.05977 0.0655

EXGR does not Granger Cause INFL 29 1.82215 0.1833

INFL does not Granger Cause EXGR 2.33617 0.1183

MS does not Granger Cause INTR 29 9.75835 0.0008

INTR does not Granger Cause MS 0.56735 0.5744

EXGR does not Granger Cause INTR 29 0.06353 0.9386

INTR does not Granger Cause EXGR 0.99658 0.3839

EXGR does not Granger Cause MS 29 1.55149 0.2324

MS does not Granger Cause EXGR 0.13971 0.8703

156
Table K:- TREND ANALYSIS GRAPH

GDP MPR
27.2 3.6

26.8 3.2

26.4 2.8

26.0 2.4

25.6 2.0

25.2 1.6
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

INFL INTR
4.4 3.6

4.0
3.2
3.6

3.2 2.8

2.8 2.4
2.4
2.0
2.0

1.6 1.6
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

MS EXGR
32 6.0

5.6
30

5.2
28
4.8
26
4.4

24
4.0

22 3.6
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

157
 APPENDIX III (USA)

Table B:- GENERALIZED METHOD OF MOMENTS (GMM) RESULT

Dependent Variable: GDP

Method: Generalized Method of Moments
Date: 09/23/17 Time: 08:34
Sample (adjusted): 1987 2016
Included observations: 28 after adjustments
Linear estimation with 1 weight update
Estimation weighting matrix: HAC (Bartlett kernel, Newey-West fixed
bandwidth = 4.0000)
Standard errors & covariance computed using estimation weighting matrix
Instrument specification: GDP(-1) MPR MPR(-1) INFL INFL(-1) INTR INTR(
-1) MS MS(-1) EXGR EXGR(-1)
Constant added to instrument list

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 1.001684 0.033121 30.24354 0.0000

MPR -0.004747 0.033745 -0.140683 0.8899
MPR(-1) -0.044892 0.033052 -1.358232 0.1932
INFL -0.002150 0.004629 -0.464507 0.6485
INFL(-1) 0.000815 0.004056 0.200945 0.8433
INTR 0.011918 0.021319 0.559022 0.5839
INTR(-1) 0.001364 0.017505 0.077892 0.9389
MS 0.229194 0.074215 3.088262 0.0071
MS(-1) -0.265897 0.083217 -3.195218 0.0056
EXGR 0.044203 0.067413 0.655707 0.5213
EXGR(-1) -0.022195 0.041615 -0.533333 0.6011
C 1.020885 0.438411 2.328600 0.0333

R-squared 0.998986 Mean dependent var 30.13307

Adjusted R-squared 0.998288 S.D. dependent var 0.229026
S.E. of regression 0.009476 Sum squared resid 0.001437
Durbin-Watson stat 2.842250 J-statistic 1.40E-33
Instrument rank 12

158
Table C:- RESIDUAL SERIES/ECM

Last updated:
09/22/17 - 10:20
Modified: 1986
2016 //
makeresids ecm

1986 NA
1987 -0.090648
1988 -0.076274
1989 -0.065705
1990 -0.007854
1991 -0.007958
1992 0.027353
1993 0.018193
1994 0.054743
1995 0.027244
1996 0.018637
1997 0.018564
1998 -0.002975
1999 0.018162
2000 0.021984
2001 -0.001614
2002 0.013114
2003 0.038178
2004 0.074868
2005 0.030030
2006 0.029276
2007 0.016063
2008 -0.001329
2009 NA
2010 0.025197
2011 -0.012823
2012 -0.034594
2013 -0.032661
2014 -0.027222
2015 -0.009104
2016 -0.060844

159
Table D:- ADF UNIT ROOT TEST (URT) RESULTS

Table D (1):- ADF URT FOR GDP @ LEVEL (NON-STATIONARY)

Null Hypothesis: GDP has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.322079 0.1719

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 10:58
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) -0.027654 0.011909 -2.322079 0.0277

C 0.858014 0.358728 2.391823 0.0237

R-squared 0.161477 Mean dependent var 0.025043

Adjusted R-squared 0.131530 S.D. dependent var 0.015981
S.E. of regression 0.014893 Akaike info criterion -5.511521
Sum squared resid 0.006210 Schwarz criterion -5.418108
Log likelihood 84.67281 Hannan-Quinn criter. -5.481637
F-statistic 5.392051 Durbin-Watson stat 1.266438
Prob(F-statistic) 0.027723

160
Table D (2):- ADF URT FOR GDP @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(GDP) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.207701 0.0298

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP,2)
Method: Least Squares
Date: 09/22/17 Time: 10:59
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(GDP(-1)) -0.551899 0.172054 -3.207701 0.0034

C 0.013372 0.005148 2.597416 0.0150

R-squared 0.275933 Mean dependent var -0.000621

Adjusted R-squared 0.249115 S.D. dependent var 0.016990
S.E. of regression 0.014723 Akaike info criterion -5.532378
Sum squared resid 0.005852 Schwarz criterion -5.438082
Log likelihood 82.21948 Hannan-Quinn criter. -5.502846
F-statistic 10.28935 Durbin-Watson stat 1.961767
Prob(F-statistic) 0.003433

Table D (3):- ADF URT FOR MPR @ LEVEL (NON-STATIONARY)

Null Hypothesis: MPR has a unit root

Exogenous: Constant
Lag Length: 2 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.135448 0.6872

Test critical values: 1% level -3.689194
5% level -2.971853

161
 10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR)
Method: Least Squares
Date: 09/22/17 Time: 11:00
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) -0.092112 0.081124 -1.135448 0.2674

D(MPR(-1)) 0.709452 0.162401 4.368512 0.0002
D(MPR(-2)) -0.473583 0.181584 -2.608068 0.0154
C 0.136180 0.150177 0.906799 0.3735

R-squared 0.487975 Mean dependent var -0.034841

Adjusted R-squared 0.423972 S.D. dependent var 0.200273
S.E. of regression 0.152000 Akaike info criterion -0.798310
Sum squared resid 0.554495 Schwarz criterion -0.607995
Log likelihood 15.17634 Hannan-Quinn criter. -0.740129
F-statistic 7.624243 Durbin-Watson stat 2.263018
Prob(F-statistic) 0.000950

Table D (4):- ADF URT FOR MPR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(MPR) has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.082998 0.0003

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR,2)
Method: Least Squares
Date: 09/22/17 Time: 11:00
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

162
 D(MPR(-1)) -0.873936 0.171933 -5.082998 0.0000
D(MPR(-1),2) 0.567081 0.162773 3.483875 0.0018
C -0.031064 0.029453 -1.054723 0.3016

R-squared 0.516615 Mean dependent var -0.001883

Adjusted R-squared 0.477945 S.D. dependent var 0.211584
S.E. of regression 0.152877 Akaike info criterion -0.817413
Sum squared resid 0.584282 Schwarz criterion -0.674677
Log likelihood 14.44379 Hannan-Quinn criter. -0.773778
F-statistic 13.35933 Durbin-Watson stat 2.344080
Prob(F-statistic) 0.000113

Table D (5):- ADF URT FOR INFL @ LEVEL (NON-STATIONARY)

Null Hypothesis: INFL has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -0.303452 0.9116

Test critical values: 1% level -3.711457
5% level -2.981038
10% level -2.629906

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL)
Method: Least Squares
Date: 09/22/17 Time: 11:01
Sample (adjusted): 1988 2016
Included observations: 26 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INFL(-1) -0.107883 0.355519 -0.303452 0.7643

D(INFL(-1)) -0.604688 0.411485 -1.469526 0.1552
C 0.011329 0.371026 0.030536 0.9759

R-squared 0.333650 Mean dependent var -0.034280

Adjusted R-squared 0.275706 S.D. dependent var 0.752659
S.E. of regression 0.640554 Akaike info criterion 2.055201
Sum squared resid 9.437126 Schwarz criterion 2.200366
Log likelihood -23.71762 Hannan-Quinn criter. 2.097003
F-statistic 5.758188 Durbin-Watson stat 2.079760
Prob(F-statistic) 0.009388

163
Table D (6):- ADF URT FOR INFL @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(INFL) has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.685236 0.0113

Test critical values: 1% level -3.737853
5% level -2.991878
10% level -2.635542

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL,2)
Method: Least Squares
Date: 09/22/17 Time: 11:04
Sample (adjusted): 1989 2016
Included observations: 24 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INFL(-1)) -1.760373 0.477683 -3.685236 0.0014

D(INFL(-1),2) -0.034755 0.407626 -0.085263 0.9329
C -0.140899 0.137644 -1.023655 0.3176

R-squared 0.742660 Mean dependent var 0.125592

Adjusted R-squared 0.718152 S.D. dependent var 1.228429
S.E. of regression 0.652165 Akaike info criterion 2.099431
Sum squared resid 8.931711 Schwarz criterion 2.246688
Log likelihood -22.19317 Hannan-Quinn criter. 2.138498
F-statistic 30.30210 Durbin-Watson stat 2.035070
Prob(F-statistic) 0.000001

164
Table D (7):- ADF URT FOR INTR @ LEVEL (NON-STATIONARY)

Null Hypothesis: INTR has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.949527 0.3062

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR)
Method: Least Squares
Date: 09/22/17 Time: 11:04
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INTR(-1) -0.180972 0.092829 -1.949527 0.0621

D(INTR(-1)) 0.391231 0.180629 2.165936 0.0397
C 0.214218 0.131225 1.632446 0.1146

R-squared 0.206437 Mean dependent var -0.032173

Adjusted R-squared 0.145393 S.D. dependent var 0.296402
S.E. of regression 0.274009 Akaike info criterion 0.346384
Sum squared resid 1.952102 Schwarz criterion 0.487829
Log likelihood -2.022575 Hannan-Quinn criter. 0.390683
F-statistic 3.381808 Durbin-Watson stat 1.869349
Prob(F-statistic) 0.049495

Table D (8):- ADF URT FOR INTR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(INTR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.815764 0.0072

Test critical values: 1% level -3.679322

165
 5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR,2)
Method: Least Squares
Date: 09/22/17 Time: 11:06
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INTR(-1)) -0.699600 0.183345 -3.815764 0.0007

C -0.021271 0.053869 -0.394872 0.6960

R-squared 0.350338 Mean dependent var 0.004118

Adjusted R-squared 0.326276 S.D. dependent var 0.350715
S.E. of regression 0.287869 Akaike info criterion 0.413853
Sum squared resid 2.237458 Schwarz criterion 0.508149
Log likelihood -4.000866 Hannan-Quinn criter. 0.443385
F-statistic 14.56005 Durbin-Watson stat 1.836076
Prob(F-statistic) 0.000719

Table D (9):- ADF URT FOR MS @ LEVEL (NON-STATIONARY)

Null Hypothesis: MS has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -0.243658 0.9217

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS)
Method: Least Squares
Date: 09/22/17 Time: 11:06
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

166
 MS(-1) -0.002597 0.010657 -0.243658 0.8094
D(MS(-1)) 0.549108 0.166982 3.288421 0.0029
C 0.100555 0.314700 0.319525 0.7519

R-squared 0.295247 Mean dependent var 0.052411

Adjusted R-squared 0.241035 S.D. dependent var 0.031964
S.E. of regression 0.027846 Akaike info criterion -4.226534
Sum squared resid 0.020161 Schwarz criterion -4.085090
Log likelihood 64.28475 Hannan-Quinn criter. -4.182236
F-statistic 5.446179 Durbin-Watson stat 1.953411
Prob(F-statistic) 0.010580

Table D (10):-ADF URT FOR MS @ FIRST DIFF (NON- STATIONARY)

Null Hypothesis: D(MS) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.823405 0.0674

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS,2)
Method: Least Squares
Date: 09/22/17 Time: 11:08
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MS(-1)) -0.457333 0.161979 -2.823405 0.0088

C 0.023915 0.009907 2.413802 0.0228

R-squared 0.227945 Mean dependent var -0.000101

Adjusted R-squared 0.199351 S.D. dependent var 0.030574
S.E. of regression 0.027357 Akaike info criterion -4.293219
Sum squared resid 0.020207 Schwarz criterion -4.198923
Log likelihood 64.25167 Hannan-Quinn criter. -4.263686
F-statistic 7.971618 Durbin-Watson stat 1.941751
Prob(F-statistic) 0.008817

167
Table D (11):-ADF URT FOR MS @ SECOND DIFF (STATIONARY)

Null Hypothesis: D(MS,2) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -6.416142 0.0000

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS,3)
Method: Least Squares
Date: 09/22/17 Time: 11:09
Sample (adjusted): 1989 2016
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MS(-1),2) -1.212564 0.188986 -6.416142 0.0000

C -0.001076 0.005777 -0.186324 0.8536

R-squared 0.612904 Mean dependent var -0.000802

Adjusted R-squared 0.598016 S.D. dependent var 0.048211
S.E. of regression 0.030567 Akaike info criterion -4.069049
Sum squared resid 0.024293 Schwarz criterion -3.973892
Log likelihood 58.96669 Hannan-Quinn criter. -4.039959
F-statistic 41.16688 Durbin-Watson stat 1.979260
Prob(F-statistic) 0.000001

Table D (12):-ADF URT FOR EXGR @ LEVEL (NON- STATIONARY)

Null Hypothesis: EXGR has a unit root

Exogenous: Constant
Lag Length: 3 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.858546 0.0636

Test critical values: 1% level -3.699871

168
 5% level -2.976263
10% level -2.627420

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXGR)
Method: Least Squares
Date: 09/22/17 Time: 11:10
Sample (adjusted): 1990 2016
Included observations: 27 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

EXGR(-1) -0.335960 0.117528 -2.858546 0.0091

D(EXGR(-1)) 0.463795 0.178506 2.598205 0.0164
D(EXGR(-2)) 0.033838 0.203837 0.166005 0.8697
D(EXGR(-3)) 0.478165 0.180175 2.653900 0.0145
C 1.574929 0.549517 2.866023 0.0090

R-squared 0.397820 Mean dependent var 0.002575

Adjusted R-squared 0.288333 S.D. dependent var 0.043899
S.E. of regression 0.037033 Akaike info criterion -3.588429
Sum squared resid 0.030172 Schwarz criterion -3.348459
Log likelihood 53.44379 Hannan-Quinn criter. -3.517073
F-statistic 3.633487 Durbin-Watson stat 2.269969
Prob(F-statistic) 0.020298

Table D (13):-ADF URT FOR EXGR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(EXGR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.297672 0.0022

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXGR,2)
Method: Least Squares
Date: 09/22/17 Time: 11:11
Sample (adjusted): 1988 2016

169
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(EXGR(-1)) -0.727518 0.169282 -4.297672 0.0002

C 0.002339 0.008011 0.292000 0.7725

R-squared 0.406202 Mean dependent var 0.004727

Adjusted R-squared 0.384209 S.D. dependent var 0.054842
S.E. of regression 0.043036 Akaike info criterion -3.387105
Sum squared resid 0.050006 Schwarz criterion -3.292809
Log likelihood 51.11302 Hannan-Quinn criter. -3.357573
F-statistic 18.46998 Durbin-Watson stat 2.053005
Prob(F-statistic) 0.000201

Table D (14):-ADF URT FOR ECM @ LEVEL (NON-STATIONARY)

Null Hypothesis: ECM has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.490772 0.1287

Test critical values: 1% level -3.699871
5% level -2.976263
10% level -2.627420

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM)
Method: Least Squares
Date: 09/22/17 Time: 11:12
Sample (adjusted): 1988 2016
Included observations: 27 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

ECM(-1) -0.310061 0.124484 -2.490772 0.0197

C 0.000835 0.004667 0.178998 0.8594

R-squared 0.198819 Mean dependent var 0.000121

Adjusted R-squared 0.166772 S.D. dependent var 0.026517
S.E. of regression 0.024205 Akaike info criterion -4.533358
Sum squared resid 0.014647 Schwarz criterion -4.437370
Log likelihood 63.20033 Hannan-Quinn criter. -4.504816
F-statistic 6.203947 Durbin-Watson stat 1.852881
Prob(F-statistic) 0.019748

170
Table D (15):-ADF URT FOR ECM @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(ECM) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.112918 0.0004

Test critical values: 1% level -3.724070
5% level -2.986225
10% level -2.632604

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM,2)
Method: Least Squares
Date: 09/22/17 Time: 11:14
Sample (adjusted): 1989 2016
Included observations: 25 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(ECM(-1)) -1.111749 0.217439 -5.112918 0.0000

C 0.001401 0.005380 0.260362 0.7969

R-squared 0.531968 Mean dependent var -0.001819

Adjusted R-squared 0.511619 S.D. dependent var 0.038224
S.E. of regression 0.026713 Akaike info criterion -4.330735
Sum squared resid 0.016412 Schwarz criterion -4.233225
Log likelihood 56.13419 Hannan-Quinn criter. -4.303690
F-statistic 26.14193 Durbin-Watson stat 1.955363
Prob(F-statistic) 0.000035

Table E (1):- ARDL BOUNDS TEST

ARDL Bounds Test

Date: 09/22/17 Time: 11:17
Sample: 1988 2016
Included observations: 26
Null Hypothesis: No long-run relationships exist

Test Statistic Value k

171
F-statistic 14.29118 6

Critical Value Bounds

Significance I0 Bound I1 Bound

10% 1.75 2.87

5% 2.04 3.24
2.5% 2.32 3.59
1% 2.66 4.05

Test Equation:
Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 11:17
Sample (adjusted): 1989 2016
Included observations: 26 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) 0.003910 0.037720 0.103659 0.9185

INFL(-1) -0.011056 0.007125 -1.551606 0.1373
INTR(-1) -0.018940 0.023249 -0.814653 0.4254
MS(-1) -0.067505 0.018262 -3.696450 0.0015
EXGR(-1) -0.002334 0.046555 -0.050143 0.9605
ECM(-2) 0.060596 0.072473 0.836115 0.4135
GDP(-1) 0.068451 0.022046 3.104886 0.0058

R-squared 0.485113 Mean dependent var 0.024430

Adjusted R-squared 0.322517 S.D. dependent var 0.016699
S.E. of regression 0.013745 Akaike info criterion -5.511484
Sum squared resid 0.003590 Schwarz criterion -5.172766
Log likelihood 78.64930 Hannan-Quinn criter. -5.413946
Durbin-Watson stat 1.371430

Table E (2):- LONG RUN RESULT

ARDL Cointegrating And Long Run Form

Dependent Variable: GDP
Selected Model: ARDL(1, 0, 0, 0, 0, 0)
Date: 10/08/17 Time: 14:23
Sample: 1986 2016
Included observations: 29

Cointegrating Form

Variable Coefficient Std. Error t-Statistic Prob.

D(MPR) 0.018805 0.027457 0.684882 0.5006

D(INFL) -0.008663 0.005640 -1.536102 0.1388
D(INTR) -0.010456 0.017343 -0.602910 0.5527

172
 D(MS) -0.033315 0.028518 -1.168219 0.2552
D(EXGR) 0.000008 0.037757 0.000218 0.9998
CointEq(-1) 0.045229 0.056787 0.796475 0.4343

Cointeq = GDP - (-0.4158*MPR + 0.1915*INFL + 0.2312*INTR + 0.7366*MS

-0.0002*EXGR + 7.9590 )

Long Run Coefficients

Variable Coefficient Std. Error t-Statistic Prob.

MPR -0.415770 0.783071 -0.530948 0.6008

INFL 0.191541 0.249884 0.766521 0.4515
INTR 0.231184 0.447721 0.516357 0.6108
MS 0.736585 0.360925 2.040829 0.0534
EXGR -0.000182 0.834883 -0.000218 0.9998
C 7.959018 10.383005 0.766543 0.4515

Table F:- LM SERIAL CORRELATION TEST

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.643029 Prob. F(2,18) 0.5374

Obs*R-squared 1.800449 Prob. Chi-Square(2) 0.4065

Test Equation:
Dependent Variable: RESID
Method: ARDL
Date: 09/22/17 Time: 11:18
Sample: 1988 2016
Included observations: 27
Presample and interior missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 0.004289 0.027408 0.156475 0.8774

MPR -0.017921 0.036971 -0.484742 0.6337
INFL 0.002770 0.006693 0.413894 0.6838
INTR 0.008836 0.021541 0.410177 0.6865
MS -0.003802 0.023438 -0.162230 0.8729
EXGR 0.000452 0.042982 0.010508 0.9917
ECM(-1) -0.016697 0.068093 -0.245209 0.8091
RESID(-1) 0.306111 0.298606 1.025134 0.3189
RESID(-2) -0.121878 0.333722 -0.365209 0.7192

R-squared 0.066683 Mean dependent var -2.21E-06

173
Adjusted R-squared -0.348124 S.D. dependent var 0.010900
S.E. of regression 0.012656 Akaike info criterion -5.640198
Sum squared resid 0.002883 Schwarz criterion -5.208253
Log likelihood 85.14268 Hannan-Quinn criter. -5.511758
Durbin-Watson stat 2.044744

Table G: - HETEROSKEDASTICITY TEST

Heteroskedasticity Test: Breusch-Pagan-Godfrey

F-statistic 0.280285 Prob. F(7,19) 0.9540

Obs*R-squared 2.527135 Prob. Chi-Square(7) 0.9250
Scaled explained SS 2.202701 Prob. Chi-Square(7) 0.9478

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 09/22/17 Time: 11:19
Sample: 1988 2016
Included observations: 27

Variable Coefficient Std. Error t-Statistic Prob.

C 0.036794 0.048573 0.757503 0.4580

GDP(-1) -0.002287 0.003064 -0.746531 0.4645
MPR 3.33E-05 0.000603 0.055251 0.9565
INFL 2.88E-05 0.000117 0.245429 0.8088
INTR 4.30E-05 0.000379 0.113397 0.9109
MS 0.000967 0.001370 0.705890 0.4888
EXGR 0.000721 0.001048 0.688298 0.4996
ECM(-1) 0.001858 0.003012 0.616944 0.5446

R-squared 0.093598 Mean dependent var 0.000114

Adjusted R-squared -0.240340 S.D. dependent var 0.000208
S.E. of regression 0.000231 Akaike info criterion -13.66331
Sum squared resid 1.02E-06 Schwarz criterion -13.27936
Log likelihood 192.4547 Hannan-Quinn criter. -13.54914
F-statistic 0.280285 Durbin-Watson stat 2.193143
Prob(F-statistic) 0.953966

174
Table H: - RAMSEY RESET TEST

Ramsey RESET Test

Equation: UNTITLED
Specification: GDP GDP(-1) MPR INFL INTR MS EXGR ECM(-1)
Omitted Variables: Squares of fitted values

Value df Probability
t-statistic 0.993516 19 0.3329
F-statistic 0.987074 (1, 19) 0.3329

F-test summary:
Mean
Sum of Sq. df Squares
Test SSR 0.000153 1 0.000153
Restricted SSR 0.003089 20 0.000154
Unrestricted SSR 0.002936 19 0.000155

Unrestricted Test Equation:

Dependent Variable: GDP
Method: ARDL
Date: 09/22/17 Time: 11:20
Sample: 1988 2016
Included observations: 27
Maximum dependent lags: 1 (Automatic selection)
Model selection method: Akaike info criterion (AIC)
Dynamic regressors (0 lag, automatic):
Fixed regressors:

Variable Coefficient Std. Error t-Statistic Prob.*

GDP(-1) 1.007250 0.023552 42.76672 0.0000

MPR 0.030242 0.031879 0.948646 0.3547
INFL -0.009690 0.006128 -1.581240 0.1303
INTR -0.018707 0.019824 -0.943645 0.3572
MS -0.091082 0.071104 -1.280963 0.2156
EXGR -0.033361 0.057485 -0.580341 0.5685
ECM(-1) -0.131979 0.165599 -0.796983 0.4353
FITTED^2 0.002909 0.002928 0.993516 0.3329

R-squared 0.997661 Mean dependent var 30.14766

Adjusted R-squared 0.996799 S.D. dependent var 0.219722
S.E. of regression 0.012432 Akaike info criterion -5.695908
Sum squared resid 0.002936 Schwarz criterion -5.311957
Log likelihood 84.89476 Hannan-Quinn criter. -5.581740
Durbin-Watson stat 1.530578

*Note: p-values and any subsequent tests do not account for model
selection.

175
Table I:- CUSUM TEST

15

10

-5

-10

-15
94 96 98 00 02 04 06 08 10 12 14 16

CUSUM 5% Significance

Table J:- NORMALITY TEST

176
12
Series: Residuals
Sample 1988 2016
10 Observations 28

8 Mean 4.61e-07
Median 0.002999
Maximum 0.014357
6 Minimum -0.027495
Std. Dev. 0.010488
Skewness -0.955992
4
Kurtosis 3.087920

2 Jarque-Bera 4.273983
Probability 0.118009
0
-0.03 -0.02 -0.01 0.00 0.01

Table K:- ENGLE GRANGER CAUSALITY TEST

Pairwise Granger Causality Tests

Date: 09/22/17 Time: 11:21
Sample: 1986 2016
Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

MPR does not Granger Cause GDP 29 2.19591 0.1331

GDP does not Granger Cause MPR 10.3340 0.0006

INFL does not Granger Cause GDP 26 2.53237 0.1035

GDP does not Granger Cause INFL 1.74106 0.1997

INTR does not Granger Cause GDP 29 0.82085 0.4520

GDP does not Granger Cause INTR 5.51882 0.0107

MS does not Granger Cause GDP 29 0.29109 0.7501

GDP does not Granger Cause MS 18.2689 2.E-05

EXGR does not Granger Cause GDP 29 0.54805 0.5851

GDP does not Granger Cause EXGR 1.92416 0.1679

INFL does not Granger Cause MPR 26 0.37942 0.6889

MPR does not Granger Cause INFL 2.08242 0.1496

INTR does not Granger Cause MPR 29 0.08430 0.9194

MPR does not Granger Cause INTR 3.51288 0.0459

MS does not Granger Cause MPR 29 16.8474 3.E-05

MPR does not Granger Cause MS 6.51056 0.0055

177
EXGR does not Granger Cause MPR 29 0.49049 0.6183
MPR does not Granger Cause EXGR 0.08506 0.9187

INTR does not Granger Cause INFL 26 1.47956 0.2505

INFL does not Granger Cause INTR 1.57555 0.2304

MS does not Granger Cause INFL 26 2.61534 0.0968

INFL does not Granger Cause MS 1.36885 0.2762

EXGR does not Granger Cause INFL 26 3.48594 0.0493

INFL does not Granger Cause EXGR 3.26703 0.0582

MS does not Granger Cause INTR 29 15.1489 6.E-05

INTR does not Granger Cause MS 1.70584 0.2029

EXGR does not Granger Cause INTR 29 0.26503 0.7694

INTR does not Granger Cause EXGR 0.39953 0.6750

EXGR does not Granger Cause MS 29 0.84346 0.4426

MS does not Granger Cause EXGR 1.73750 0.1974

Table L:- TREND ANALYSIS GRAPH

178
 GDP MPR
30.6 2.4

2.2
30.4
2.0
30.2 1.8

30.0 1.6

1.4
29.8
1.2

29.6 1.0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

INFL INTR
2 2.0

1 1.6

0 1.2

-1 0.8

-2 0.4

-3 0.0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

MS EXGR
30.8 4.85

4.80
30.4

4.75
30.0
4.70
29.6
4.65

29.2
4.60

28.8 4.55
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

179
 APPENDIX IV (UK)

Table B:- GENERALIZED METHOD OF MOMENTS (GMM) RESULT

Dependent Variable: GDP

Method: Generalized Method of Moments
Date: 09/23/17 Time: 08:40
Sample (adjusted): 1990 2008
Included observations: 17 after adjustments
Linear estimation with 1 weight update
Estimation weighting matrix: HAC (Bartlett kernel, Newey-West fixed
bandwidth = 3.0000)
Standard errors & covariance computed using estimation weighting matrix
Instrument specification: GDP(-1) MPR MPR(-1) INFL INFL(-1) INTR INTR(
-1) MS MS(-1) EXGR EXGR(-1)
Constant added to instrument list

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) 1.122529 0.338671 3.314513 0.0211

MPR -0.013038 0.043772 -0.297868 0.7778
MPR(-1) -0.038857 0.030101 -1.290861 0.2532
INFL -0.002953 0.010954 -0.269628 0.7982
INFL(-1) 0.003916 0.007993 0.489972 0.6449
INTR 0.012237 0.007611 1.607793 0.1688
INTR(-1) 0.002220 0.009531 0.232951 0.8250
MS 0.013668 0.052918 0.258290 0.8065
MS(-1) -0.086118 0.031568 -2.728050 0.0414
EXGR 0.086455 0.089642 0.964453 0.3791
EXGR(-1) 0.083411 0.082760 1.007860 0.3598
C -2.178914 7.551061 -0.288557 0.7845

R-squared 0.999179 Mean dependent var 28.33858

Adjusted R-squared 0.997371 S.D. dependent var 0.158624
S.E. of regression 0.008133 Sum squared resid 0.000331
Durbin-Watson stat 2.906495 J-statistic 4.11E-36
Instrument rank 12

180
Table C:- RESIDUAL SERIES/ECM

Last updated:
09/22/17 - 11:56
Modified: 1986
2016 //
makeresids ecm

1986 NA
1987 NA
1988 NA
1989 -0.005251
1990 0.014122
1991 -0.015383
1992 -0.000387
1993 0.011155
1994 0.002758
1995 NA
1996 NA
1997 -3.42E-05
1998 -0.012213
1999 -0.002983
2000 0.005226
2001 -0.000151
2002 0.004016
2003 -0.014644
2004 -0.009114
2005 0.005102
2006 0.015277
2007 0.011097
2008 -0.008593
2009 NA
2010 NA
2011 NA
2012 NA
2013 NA
2014 NA
2015 NA
2016 NA

181
Table D:- ADF UNIT ROOT TEST (URT) RESULTS

Table D (1):- ADF URT FOR GDP @ LEVEL (NON-STATIONARY)

Null Hypothesis: GDP has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -1.801552 0.3726

Test critical values: 1% level -3.670170
5% level -2.963972
10% level -2.621007

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 11:58
Sample (adjusted): 1987 2016
Included observations: 30 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) -0.031141 0.017286 -1.801552 0.0824

C 0.904355 0.489941 1.845846 0.0755

R-squared 0.103874 Mean dependent var 0.021721

Adjusted R-squared 0.071869 S.D. dependent var 0.018764
S.E. of regression 0.018077 Akaike info criterion -5.124009
Sum squared resid 0.009150 Schwarz criterion -5.030596
Log likelihood 78.86013 Hannan-Quinn criter. -5.094125
F-statistic 3.245591 Durbin-Watson stat 1.027233
Prob(F-statistic) 0.082394

182
Table D (2):- ADF URT FOR GDP @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(GDP) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.295607 0.0244

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP,2)
Method: Least Squares
Date: 09/22/17 Time: 11:58
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(GDP(-1)) -0.522797 0.158634 -3.295607 0.0028

C 0.010242 0.004568 2.242140 0.0334

R-squared 0.286866 Mean dependent var -0.001183

Adjusted R-squared 0.260453 S.D. dependent var 0.018626
S.E. of regression 0.016018 Akaike info criterion -5.363787
Sum squared resid 0.006927 Schwarz criterion -5.269491
Log likelihood 79.77491 Hannan-Quinn criter. -5.334254
F-statistic 10.86103 Durbin-Watson stat 1.828709
Prob(F-statistic) 0.002751

Table D (3):- ADF URT FOR MPR @ LEVEL (NON-STATIONARY)

Null Hypothesis: MPR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -0.244817 0.9213

Test critical values: 1% level -3.689194
5% level -2.971853

183
 10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR)
Method: Least Squares
Date: 09/22/17 Time: 12:11
Sample (adjusted): 1987 2014
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) -0.018421 0.075242 -0.244817 0.8085

C -0.083477 0.133758 -0.624090 0.5380

R-squared 0.002300 Mean dependent var -0.110073

Adjusted R-squared -0.036073 S.D. dependent var 0.405686
S.E. of regression 0.412939 Akaike info criterion 1.137714
Sum squared resid 4.433477 Schwarz criterion 1.232871
Log likelihood -13.92799 Hannan-Quinn criter. 1.166804
F-statistic 0.059935 Durbin-Watson stat 1.704710
Prob(F-statistic) 0.808520

Table D (4):- ADF URT FOR MPR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(MPR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.368365 0.0020

Test critical values: 1% level -3.699871
5% level -2.976263
10% level -2.627420

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MPR,2)
Method: Least Squares
Date: 09/22/17 Time: 12:12
Sample (adjusted): 1988 2014
Included observations: 27 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MPR(-1)) -0.867180 0.198514 -4.368365 0.0002

C -0.094799 0.083552 -1.134612 0.2673

184
R-squared 0.432883 Mean dependent var 0.004189
Adjusted R-squared 0.410198 S.D. dependent var 0.544120
S.E. of regression 0.417876 Akaike info criterion 1.163925
Sum squared resid 4.365517 Schwarz criterion 1.259913
Log likelihood -13.71299 Hannan-Quinn criter. 1.192467
F-statistic 19.08262 Durbin-Watson stat 1.942798
Prob(F-statistic) 0.000192

Table D (5):- ADF URT FOR INFL @ LEVEL (NON-STATIONARY)

Null Hypothesis: INFL has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.720764 0.0836

Test critical values: 1% level -3.699871
5% level -2.976263
10% level -2.627420

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL)
Method: Least Squares
Date: 09/22/17 Time: 12:13
Sample (adjusted): 1990 2016
Included observations: 27 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INFL(-1) -0.463131 0.170221 -2.720764 0.0117

C 0.251550 0.195673 1.285562 0.2104

R-squared 0.228456 Mean dependent var -0.077764

Adjusted R-squared 0.197594 S.D. dependent var 0.891847
S.E. of regression 0.798891 Akaike info criterion 2.460002
Sum squared resid 15.95565 Schwarz criterion 2.555989
Log likelihood -31.21002 Hannan-Quinn criter. 2.488544
F-statistic 7.402556 Durbin-Watson stat 2.015920
Prob(F-statistic) 0.011682

185
Table D (6):- ADF URT FOR INFL @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(INFL) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=6)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -6.753291 0.0000

Test critical values: 1% level -3.711457
5% level -2.981038
10% level -2.629906

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL,2)
Method: Least Squares
Date: 09/22/17 Time: 12:15
Sample (adjusted): 1991 2016
Included observations: 26 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INFL(-1)) -1.533289 0.227043 -6.753291 0.0000

C -0.187161 0.168574 -1.110260 0.2779

R-squared 0.655207 Mean dependent var 0.087131

Adjusted R-squared 0.640840 S.D. dependent var 1.392027
S.E. of regression 0.834241 Akaike info criterion 2.549214
Sum squared resid 16.70299 Schwarz criterion 2.645991
Log likelihood -31.13979 Hannan-Quinn criter. 2.577083
F-statistic 45.60694 Durbin-Watson stat 1.988087
Prob(F-statistic) 0.000001

Table D (7):- ADF URT FOR INTR @ LEVEL (NON-STATIONARY)

Null Hypothesis: INTR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.183440 0.2176

Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413

186
*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR)
Method: Least Squares
Date: 09/22/17 Time: 12:16
Sample (adjusted): 1987 2008
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

INTR(-1) -0.380832 0.174418 -2.183440 0.0425

C 0.443979 0.230892 1.922889 0.0705

R-squared 0.209396 Mean dependent var -0.022396

Adjusted R-squared 0.165474 S.D. dependent var 0.429226
S.E. of regression 0.392108 Akaike info criterion 1.060081
Sum squared resid 2.767477 Schwarz criterion 1.159654
Log likelihood -8.600808 Hannan-Quinn criter. 1.079519
F-statistic 4.767409 Durbin-Watson stat 1.582685
Prob(F-statistic) 0.042486

Table D (8):- ADF URT FOR INTR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(INTR) has a unit root

Exogenous: Constant
Lag Length: 4 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.886822 0.0183

Test critical values: 1% level -4.297073
5% level -3.212696
10% level -2.747676

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 10

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR,2)
Method: Least Squares
Date: 09/22/17 Time: 12:16
Sample (adjusted): 1992 2008
Included observations: 10 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

187
 D(INTR(-1)) -3.194608 0.821908 -3.886822 0.0177
D(INTR(-1),2) 1.526720 0.691366 2.208267 0.0918
D(INTR(-2),2) 1.435635 0.600165 2.392068 0.0750
D(INTR(-3),2) 0.949764 0.358726 2.647599 0.0571
D(INTR(-4),2) 0.912338 0.334792 2.725088 0.0527
C -0.266176 0.124484 -2.138242 0.0993

R-squared 0.913607 Mean dependent var -0.006067

Adjusted R-squared 0.805616 S.D. dependent var 0.751486
S.E. of regression 0.331323 Akaike info criterion 0.912262
Sum squared resid 0.439099 Schwarz criterion 1.093813
Log likelihood 1.438690 Hannan-Quinn criter. 0.713101
F-statistic 8.460023 Durbin-Watson stat 3.028419
Prob(F-statistic) 0.029879

Table D (9):- ADF URT FOR MS @ LEVEL (NON-STATIONARY)

Null Hypothesis: MS has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -0.603857 0.8549

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS)
Method: Least Squares
Date: 09/22/17 Time: 12:17
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MS(-1) -0.016245 0.026903 -0.603857 0.5512

D(MS(-1)) 0.233784 0.132946 1.758495 0.0904
C 0.501259 0.748490 0.669695 0.5090

R-squared 0.141173 Mean dependent var 0.072863

Adjusted R-squared 0.075110 S.D. dependent var 0.105148
S.E. of regression 0.101123 Akaike info criterion -1.647269
Sum squared resid 0.265870 Schwarz criterion -1.505825
Log likelihood 26.88540 Hannan-Quinn criter. -1.602971
F-statistic 2.136933 Durbin-Watson stat 1.807157

188
Prob(F-statistic) 0.138284

Table D (10):-ADF URT FOR MS @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(MS) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.870081 0.0000

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(MS,2)
Method: Least Squares
Date: 09/22/17 Time: 12:17
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(MS(-1)) -0.745792 0.127050 -5.870081 0.0000

C 0.049478 0.021930 2.256193 0.0324

R-squared 0.560674 Mean dependent var -0.019129

Adjusted R-squared 0.544403 S.D. dependent var 0.148043
S.E. of regression 0.099926 Akaike info criterion -1.702308
Sum squared resid 0.269599 Schwarz criterion -1.608011
Log likelihood 26.68346 Hannan-Quinn criter. -1.672775
F-statistic 34.45785 Durbin-Watson stat 1.852881
Prob(F-statistic) 0.000003

Table D (11):-ADF URT FOR EXGR @ LEVEL (NON-STATIONARY)

Null Hypothesis: EXGR has a unit root

Exogenous: Constant
Lag Length: 1 (Automatic - based on SIC, maxlag=7)

189
 t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -2.841615 0.0649

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXGR)
Method: Least Squares
Date: 09/22/17 Time: 12:18
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

EXGR(-1) -0.356314 0.125391 -2.841615 0.0086

D(EXGR(-1)) 0.358392 0.189697 1.889288 0.0701
C 1.669444 0.586416 2.846859 0.0085

R-squared 0.258209 Mean dependent var 0.004839

Adjusted R-squared 0.201149 S.D. dependent var 0.064552
S.E. of regression 0.057695 Akaike info criterion -2.769586
Sum squared resid 0.086547 Schwarz criterion -2.628141
Log likelihood 43.15899 Hannan-Quinn criter. -2.725287
F-statistic 4.525161 Durbin-Watson stat 1.881547
Prob(F-statistic) 0.020590

Table D (12):-ADF URT FOR EXGR @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(EXGR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=7)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.106185 0.0035

Test critical values: 1% level -3.679322
5% level -2.967767
10% level -2.622989

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXGR,2)

190
Method: Least Squares
Date: 09/22/17 Time: 12:21
Sample (adjusted): 1988 2016
Included observations: 29 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(EXGR(-1)) -0.823644 0.200586 -4.106185 0.0003

C 0.003359 0.012153 0.276358 0.7844

R-squared 0.384416 Mean dependent var -0.003554

Adjusted R-squared 0.361616 S.D. dependent var 0.081121
S.E. of regression 0.064815 Akaike info criterion -2.568090
Sum squared resid 0.113426 Schwarz criterion -2.473794
Log likelihood 39.23731 Hannan-Quinn criter. -2.538558
F-statistic 16.86076 Durbin-Watson stat 1.760106
Prob(F-statistic) 0.000334

Table D (13):-ADF URT FOR ECM @ LEVEL (NON-STATIONARY)

Null Hypothesis: ECM has a unit root

Exogenous: Constant
Lag Length: 3 (Automatic - based on SIC, maxlag=3)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -3.187171 0.0519

Test critical values: 1% level -4.297073
5% level -3.212696
10% level -2.747676

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 10

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM)
Method: Least Squares
Date: 09/22/17 Time: 12:21
Sample (adjusted): 1993 2008
Included observations: 10 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

ECM(-1) -2.876484 0.902520 -3.187171 0.0243

D(ECM(-1)) 2.027646 0.734110 2.762045 0.0397
D(ECM(-2)) 1.169764 0.603963 1.936813 0.1105
D(ECM(-3)) 0.603712 0.360682 1.673806 0.1550
C -0.002587 0.002731 -0.947361 0.3870

191
R-squared 0.796973 Mean dependent var -0.001067
Adjusted R-squared 0.634551 S.D. dependent var 0.012148
S.E. of regression 0.007344 Akaike info criterion -6.683059
Sum squared resid 0.000270 Schwarz criterion -6.531767
Log likelihood 38.41530 Hannan-Quinn criter. -6.849027
F-statistic 4.906807 Durbin-Watson stat 2.144068
Prob(F-statistic) 0.055579

Table D (14):-ADF URT FOR ECM @ FIRST DIFF (STATIONARY)

Null Hypothesis: D(ECM) has a unit root

Exogenous: Constant
Lag Length: 3 (Automatic - based on SIC, maxlag=3)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -7.296096 0.0006

Test critical values: 1% level -4.582648
5% level -3.320969
10% level -2.801384

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 8

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(ECM,2)
Method: Least Squares
Date: 09/22/17 Time: 12:23
Sample (adjusted): 1994 2008
Included observations: 8 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(ECM(-1)) -4.451779 0.610159 -7.296096 0.0053

D(ECM(-1),2) 2.852733 0.477165 5.978503 0.0094
D(ECM(-2),2) 2.115188 0.383415 5.516709 0.0117
D(ECM(-3),2) 1.069791 0.218135 4.904271 0.0162
C 0.002049 0.002145 0.955117 0.4100

R-squared 0.955496 Mean dependent var -0.004282

Adjusted R-squared 0.896158 S.D. dependent var 0.016847
S.E. of regression 0.005429 Akaike info criterion -7.325052
Sum squared resid 8.84E-05 Schwarz criterion -7.275401
Log likelihood 34.30021 Hannan-Quinn criter. -7.659928
F-statistic 16.10249 Durbin-Watson stat 2.806289
Prob(F-statistic) 0.022844

192
Table E (1):- ARDL BOUNDS TEST

ARDL Bounds Test

Date: 09/22/17 Time: 12:24
Sample: 1990 2008
Included observations: 15
Null Hypothesis: No long-run relationships exist

Test Statistic Value k

F-statistic 18.51191 6

Critical Value Bounds

Significance I0 Bound I1 Bound

10% 1.75 2.87

5% 2.04 3.24
2.5% 2.32 3.59
1% 2.66 4.05

Test Equation:
Dependent Variable: D(GDP)
Method: Least Squares
Date: 09/22/17 Time: 12:24
Sample (adjusted): 1991 2008
Included observations: 15 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

MPR(-1) -0.046956 0.025312 -1.855070 0.1007

INFL(-1) -0.002232 0.009504 -0.234798 0.8203
INTR(-1) 0.004373 0.011472 0.381205 0.7130
MS(-1) -0.042324 0.022557 -1.876296 0.0975
EXGR(-1) 0.125899 0.116938 1.076633 0.3130
ECM(-2) -0.320305 0.290330 -1.103246 0.3020
GDP(-1) 0.023791 0.007693 3.092546 0.0148

R-squared 0.804177 Mean dependent var 0.022108

Adjusted R-squared 0.657309 S.D. dependent var 0.014872
S.E. of regression 0.008706 Akaike info criterion -6.344892
Sum squared resid 0.000606 Schwarz criterion -6.014469
Log likelihood 54.58669 Hannan-Quinn criter. -6.348412
Durbin-Watson stat 2.254076

193
Table E (2):- LONG RUN RESULT

ARDL Cointegrating And Long Run Form

Dependent Variable: GDP
Selected Model: ARDL(1, 0, 0, 0, 0, 0)
Date: 10/08/17 Time: 14:27
Sample: 1986 2016
Included observations: 19

Cointegrating Form

Variable Coefficient Std. Error t-Statistic Prob.

D(MPR) 0.023723 0.025607 0.926417 0.3725

D(INFL) -0.027882 0.009041 -3.083765 0.0095
D(INTR) -0.005571 0.009496 -0.586639 0.5683
D(MS) -0.012079 0.034414 -0.350988 0.7317
D(EXGR) -0.034781 0.060095 -0.578761 0.5735
CointEq(-1) 0.052142 0.162409 0.321054 0.7537

Cointeq = GDP - (-0.4550*MPR + 0.5347*INFL + 0.1068*INTR + 0.2317*MS

+ 0.6670*EXGR + 18.6305 )

Long Run Coefficients

Variable Coefficient Std. Error t-Statistic Prob.

MPR -0.454971 1.148627 -0.396099 0.6990

INFL 0.534722 1.597021 0.334825 0.7435
INTR 0.106834 0.401268 0.266241 0.7946
MS 0.231656 0.223696 1.035582 0.3208
EXGR 0.667035 1.942477 0.343394 0.7372
C 18.630471 11.668807 1.596605 0.1363

Table F:- LM SERIAL CORRELATION TEST

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.207913 Prob. F(2,7) 0.8171

Obs*R-squared 0.897166 Prob. Chi-Square(2) 0.6385

Test Equation:
Dependent Variable: RESID
Method: ARDL
Date: 09/22/17 Time: 12:26
Sample: 1990 2008
Included observations: 16

194
Presample and interior missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

GDP(-1) -0.006876 0.016733 -0.410938 0.6934

MPR -0.004177 0.031534 -0.132461 0.8983
INFL 0.003145 0.013100 0.240090 0.8171
INTR 0.006394 0.017784 0.359552 0.7298
MS 0.006133 0.019950 0.307412 0.7675
EXGR 0.004886 0.086092 0.056748 0.9563
ECM(-1) -0.010030 0.351632 -0.028524 0.9780
RESID(-1) -0.081765 0.661036 -0.123692 0.9050
RESID(-2) -0.473027 0.761886 -0.620863 0.5544

R-squared 0.056073 Mean dependent var -6.73E-07

Adjusted R-squared -1.022701 S.D. dependent var 0.007629
S.E. of regression 0.010850 Akaike info criterion -5.910978
Sum squared resid 0.000824 Schwarz criterion -5.476396
Log likelihood 56.28782 Hannan-Quinn criter. -5.888723
Durbin-Watson stat 1.957629

Table G: - HETEROSKEDASTICITY TEST

Heteroskedasticity Test: Breusch-Pagan-Godfrey

F-statistic 1.423832 Prob. F(7,8) 0.3141

Obs*R-squared 8.875758 Prob. Chi-Square(7) 0.2617
Scaled explained SS 0.727097 Prob. Chi-Square(7) 0.9981

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 09/22/17 Time: 12:27
Sample: 1990 2008
Included observations: 16

Variable Coefficient Std. Error t-Statistic Prob.

C -0.008511 0.015039 -0.565956 0.5869

GDP(-1) 0.000359 0.000678 0.529254 0.6110
MPR 4.55E-05 0.000132 0.345055 0.7390
INFL 2.96E-05 4.50E-05 0.658797 0.5285
INTR 1.14E-06 4.73E-05 0.024225 0.9813
MS -3.48E-05 0.000149 -0.233361 0.8213
EXGR -0.000155 0.000304 -0.511498 0.6228
ECM(-1) 0.000833 0.001188 0.701211 0.5031

195
R-squared 0.554735 Mean dependent var 5.46E-05
Adjusted R-squared 0.165128 S.D. dependent var 4.06E-05
S.E. of regression 3.71E-05 Akaike info criterion -17.26165
Sum squared resid 1.10E-08 Schwarz criterion -16.87535
Log likelihood 146.0932 Hannan-Quinn criter. -17.24187
F-statistic 1.423832 Durbin-Watson stat 2.777866
Prob(F-statistic) 0.314147

Table H: - RAMSEY RESET TEST

Ramsey RESET Test

Equation: UNTITLED
Specification: GDP GDP(-1) MPR INFL INTR MS EXGR ECM(-1)
Omitted Variables: Squares of fitted values

Value df Probability
t-statistic 0.423801 8 0.6829
F-statistic 0.179607 (1, 8) 0.6829

F-test summary:
Mean
Sum of Sq. df Squares
Test SSR 1.92E-05 1 1.92E-05
Restricted SSR 0.000873 9 9.70E-05
Unrestricted SSR 0.000854 8 0.000107

Unrestricted Test Equation:

Dependent Variable: GDP
Method: ARDL
Date: 09/22/17 Time: 12:27
Sample: 1990 2008
Included observations: 16
Maximum dependent lags: 1 (Automatic selection)
Model selection method: Akaike info criterion (AIC)
Dynamic regressors (0 lag, automatic):
Fixed regressors:

Variable Coefficient Std. Error t-Statistic Prob.*

GDP(-1) 0.963372 0.109064 8.833112 0.0000

MPR 0.007854 0.036888 0.212924 0.8367
INFL -0.020923 0.011288 -1.853585 0.1009
INTR -0.000507 0.013179 -0.038472 0.9703
MS -0.024856 0.038601 -0.643914 0.5377
EXGR -0.001991 0.085231 -0.023355 0.9819

196
 ECM(-1) 0.132035 0.323335 0.408354 0.6937
FITTED^2 0.002187 0.005160 0.423801 0.6829

R-squared 0.997845 Mean dependent var 28.34343

Adjusted R-squared 0.995959 S.D. dependent var 0.162520
S.E. of regression 0.010331 Akaike info criterion -6.000474
Sum squared resid 0.000854 Schwarz criterion -5.614179
Log likelihood 56.00379 Hannan-Quinn criter. -5.980692
Durbin-Watson stat 1.663228

*Note: p-values and any subsequent tests do not account for model
selection.

Table I:- CUSUM TEST

12

-4

-8

-12
96 97 98 99 00 01 02 03 04 05 06 07 08

CUSUM 5% Significance

Table J:- NORMALITY TEST

197
4
Series: Residuals
Sample 1990 2008
Observations 16
3
Mean -6.73e-07
Median 0.001472
Maximum 0.010018
2 Minimum -0.011476
Std. Dev. 0.007629
Skewness -0.091813
Kurtosis 1.517778
1
Jarque-Bera 1.487134
Probability 0.475415

0
-0.010 -0.005 0.000 0.005 0.010

Table K:- ENGLE GRANGER CAUSALITY TEST

Pairwise Granger Causality Tests

Date: 09/22/17 Time: 12:31
Sample: 1986 2016
Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

MPR does not Granger Cause GDP 27 8.02204 0.0024

GDP does not Granger Cause MPR 6.84234 0.0049

INFL does not Granger Cause GDP 26 0.85814 0.4383

GDP does not Granger Cause INFL 1.58696 0.2281

INTR does not Granger Cause GDP 18 1.46729 0.2663

GDP does not Granger Cause INTR 3.04419 0.0823

MS does not Granger Cause GDP 29 0.38526 0.6844

GDP does not Granger Cause MS 9.17098 0.0011

EXGR does not Granger Cause GDP 29 0.67054 0.5208

GDP does not Granger Cause EXGR 0.92691 0.4095

198
INFL does not Granger Cause MPR 24 1.35752 0.2811
MPR does not Granger Cause INFL 0.30033 0.7440

INTR does not Granger Cause MPR 18 2.32822 0.1367

MPR does not Granger Cause INTR 1.88878 0.1905

MS does not Granger Cause MPR 27 3.22232 0.0592

MPR does not Granger Cause MS 0.11718 0.8900

EXGR does not Granger Cause MPR 27 6.20330 0.0073

MPR does not Granger Cause EXGR 0.59490 0.5603

INTR does not Granger Cause INFL 15 1.58350 0.2527

INFL does not Granger Cause INTR 1.49209 0.2710

MS does not Granger Cause INFL 26 0.66488 0.5248

INFL does not Granger Cause MS 2.49366 0.1067

EXGR does not Granger Cause INFL 26 0.20813 0.8138

INFL does not Granger Cause EXGR 2.71334 0.0895

MS does not Granger Cause INTR 18 2.39602 0.1301

INTR does not Granger Cause MS 1.47323 0.2650

EXGR does not Granger Cause INTR 18 3.91429 0.0467

INTR does not Granger Cause EXGR 1.65708 0.2285

EXGR does not Granger Cause MS 29 0.19917 0.8207

MS does not Granger Cause EXGR 1.37075 0.2731

199
 Table L:- TREND ANALYSIS GRAPH

GDP MPR
28.8 3

28.6
2

28.4
1
28.2

0
28.0

27.8 -1
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

INFL INTR
3 2.0

2
1.6
1

0 1.2

-1 0.8
-2
0.4
-3

-4 0.0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

MS EXGR
29.0 4.9

28.5
4.8
28.0

27.5
4.7
27.0

26.5
4.6
26.0

25.5 4.5
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015

200

View publication stats