University of Cape Coast [Link]

gh/xmlui

UNIVERSITY OF CAPE COAST

EMPIRICAL ANALYSIS OF THE EFFECT OF CRUDE OIL PRICE

VOLATILITY ON INDUSTRIAL OUTPUT IN GHANA

MATILDA MAKAFUI YABANI

2023

Digitized by Sam Jonah Library

University of Cape Coast [Link]

UNIVERSITY OF CAPE COAST

EMPIRICAL ANALYSIS OF THE EFFFECT OF CRUDE OIL PRICE

VOLATILITY ON INDUSTRIAL OUTPUT IN GHANA

BY

MATILDA MAKAFUI YABANI

Thesis submitted to the Institute for Oil and Gas Studies of the Faculty of
Social Sciences, College of Humanities and Legal Studies, University of Cape
Coast, in partial fulfillment of the requirements for the award of Master of
Philosophy degree in Oil and Gas Resource Management

APRIL 2023

Digitized by Sam Jonah Library

University of Cape Coast [Link]

DECLARATION

Candidate’s Declaration

I hereby declare that this thesis is the result of my original research and that no

part of it has been presented for another degree at this university or elsewhere.

Candidate’s Signature: .............................................. Date: ...........................

Name: Matilda Makafui Yabani

Supervisor’s Declaration

I hereby declare that the preparation and presentation of the thesis was

supervised in accordance with the guidelines on supervision of thesis laid

down by the University of Cape Coast.

Supervisor’s Signature: .................................... Date: .........................

Name: Prof. Omowumi O. Iledare

ii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

ABSTRACT

The industrial sector of Ghana is the second largest consumer of refined crude

oil products. The industrial sector's output has exhibited significant volatility,

with a GDP share of 19% in 2009 and 28.26% in 2021. A rising number of

worldwide economies are particularly worried about the effect of crude oil

price volatility on economic growth. Hence, this study investigated the impact

of crude oil price volatility on industrial sector output in Ghana. The study

employed monthly data from January 2001 to December 2020. The quantile

regression, the causality in quantiles and bi-wavelet approaches were used to

examine the relationship between crude oil price volatility and industrial

sector output. The study’s empirical model results showed that there was a

negative asymmetric impact of crude oil price volatility on industrial sector

output at normal and boom economic conditions. Also, the study revealed

significant causality between crude oil price volatility and industrial sector

output mostly at stress and normal economic conditions of the industrial

sector. The study further found negative significant short-term co-movements

between crude oil price volatility and industrial sector output across time and

frequency. The study recommended that policymakers should consider

implementing petroleum price policies that can mitigate the negative effects of

crude oil price volatility on the industrial sector output This includes

establishing a Price Stabilization Fund which allows the government to save

windfall gains during periods of high oil prices and release these funds to

offset rising costs when prices fall. This helps stabilize domestic petroleum

product prices, ensuring that industrial producers are shielded from sudden

crude oil price fluctuations.

iii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

KEY WORDS

Bi-wavelet

Conditional Causality in quantile

Crude oil price

Industrial sector output

Quantile regression

Volatility

iv

Digitized by Sam Jonah Library

University of Cape Coast [Link]

ACKNOWLEDGEMENTS

I express my gratitude to my supervisor Prof. Prof. Omowumi O.

Iledare for his time, encouragement, guidance and advice through all the

stages of this study.

I extend my sincere appreciation to my family for encouragement and

moral support towards my education.

Digitized by Sam Jonah Library

University of Cape Coast [Link]

DEDICATION

To my family: Emmanuel, Elizabeth, Perfect, and Samuel Yabani.

vi

Digitized by Sam Jonah Library

University of Cape Coast [Link]

TABLE OF CONTENTS

Page

DECLARATION ii

ABSTRACT iii

KEY WORDS iv

ACKNOWLEDGEMENT v

DEDICATION vi

TABLE OF CONTENTS vii

LIST OF TABLES x

LIST OF ACRONYMS xiii

CHAPTER ONE: INTRODUCTION

Background to the Study 2

Statement of the Problem 8

Purpose of the Study 16

Research Objectives 16

Research Questions 16

Significance of the Study 16

Delimitations 17

Limitations 17

Organisation of the Study 18

CHAPTER TWO: LITERATURE REVIEW

Introduction 19

Theoretical Review 19

Conceptual Review 23

Conceptual Framework 27

vii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Empirical Review 30

Chapter Summary 42

CHAPTER THREE: RESEARCH METHODS

Introduction 51

Research Philosophical Perspectives or Paradigm 51

Research Design 52

Research Approach 53

Data Sources 54

Theoretical Model Specification 54

Empirical Model Specification 55

Estimation Techniques 55

Variable Descriptions and Sources of Data 63

Data Processing and Analysis 65

Chapter Summary 67

CHAPTER FOUR: RESULTS AND DISCUSSION

Introduction 68

Descriptive Statistics 68

Correlation Matrix 75

The Asymmetric Relationship between Realised Crude Oil Price Volatility

and Industrial Sector Output 76

The Conditional Causality of Realised Crude Oil Price Volatility to

Industrial Sector Output 91

The Responsiveness of The Industrial Sector Output to Realised Crude Oil

Price Volatility Across Time and Frequency 99

Chapter Summary 109

viii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

CHAPTER FIVE: SUMMARY, CONCLUSIONS AND

RECOMMENDATIONS

Introduction 110

Summary of Findings 111

Conclusions 112

Recommendations 114

Suggestions for Further Studies 116

REFERENCES 117

APPENDICES 139

APPENDIX A 139

APPENDIX B 142

ix

Digitized by Sam Jonah Library

University of Cape Coast [Link]

LIST OF TABLES

Table Page

1 Summary and Relevance 48

2 Description of Variables, Units, and Sources of Data 64

3 Descriptive Statistics 71

4 Quantile Regression Estimates (ROVX on CONS) 77

5 Quantile Regression Estimates (ROVX on ELECT) 79

6 Quantile Regression Estimates (ROVX on MFG) 81

7 Quantile Regression Estimates (ROVX on MQ) 83

8 Quantile Regression Estimates (ROVX on WS) 85

9 Quantile Regression Estimates (ROVX on ISO) 87

Digitized by Sam Jonah Library

University of Cape Coast [Link]

LIST OF FIGURES
Figure Page

1 The Trend in Average Monthly Brent Crude Oil Prices 3

2 The Trend in Crude Oil Import in Ghana 5

3 Growth rate of GDP for Agriculture Sector, Industrial Sector and

Services Sector. 6

4 The Trend in Petroleum Product Consumption by Sectors (Ktoe) 10

5 Growth Rate of the Sub-sectors of the Industrial Sector 26

6 Conceptual Framework 30

7 Time series plots of ROVX and Industrial Sector Output 74

8 Quantile Causality in Mean from Realised Crude Oil Volatility to the

Construction Sub-sector Output 92

9 Quantile Causality in Mean from Realised Crude Oil Price Volatility

to the Electricity Sub-sector Output 93

10 Quantile Causality in Mean from Realised Crude Oil Price Volatility

to the Manufacturing Sub-sector Output 94

11 Quantile Causality in Mean from Realised Crude Oil Price Volatility

to the Mining and Quarrying Sub-sector Output 95

12 Quantile Causality in Mean from Realised Crude Oil Price Volatility

to the Water and Sewerage Sub-sector Output 96

13 Quantile Causality in Mean from Realised Crude Oil Price

Volatility to Industrial Sector Output 97

14 Co-movements between Realised Crude Oil Volatility and

Construction Sub-sector Output 100

15 Co-movements between Realised Crude Oil Volatility and

Electricity Sub-sector Output 101

xi

Digitized by Sam Jonah Library

University of Cape Coast [Link]

16 Co-movements between Realised Crude Oil Price Volatility and

Manufacturing Sub-sector Output 103

17 Co-movements between Realised Crude Oil Price Volatility and

Mining and Quarrying Sub-sector Output 104

18 Co-movements between Realised Crude Oil Price Volatility and

Water and Sewerage Sub-sector Output 105

19 Co-movements between Realised Crude Oil Price Volatility and

Industrial Sector Output 106

xii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

LIST OF ACRONYMS

BoG Bank of Ghana

CONS Construction

EIA Energy Information Administration

ELECT Electricity

GDP Gross Domestic Product

GSS Ghana Statistical Service

IEA International Energy Agency

IMF International Monetary Fund

ISO Industrial Sector Output

MFG Manufacturing

MQ Mining and Quarrying

OPEC Organization of Petroleum Exporting Countries

OPV Oil Price Volatility

ROVX Realised Crude Oil Price Volatility

WS Water and Sewerage

xiii

Digitized by Sam Jonah Library

University of Cape Coast [Link]

CHAPTER ONE

INTRODUCTION

Energy stands as a cornerstone of a nation's economic growth.

Conferring to the Energy Information Administration (2021), petroleum is

among the primary sources of energy that is utilized in the industrial sectors.

In addition to a number of other important aspects, one of the most significant

components in the production process of the global economy is crude oil. A

rising number of worldwide economies are particularly worried about

variations in crude oil price and how these may affect economic growth. These

concern stems from the possibility that crude oil price changes may inhibit

performance asymmetrically (Boateng, Adam & Owusu Junior, 2021; Mo,

Chen, Nie & Jiang, 2019) and across time and frequency (Asafo-Adjei, Adam

& Darkwa, 2021).

The largest economic sector of Ghana is typically the services sector

(Ghana Statistical Service 2021). This sector encompasses a diverse variety of

activities such as finance, tourism, telecommunications, retail, and more. It is

important to Ghana's economy, accounting for a sizable amount of the nation's

Gross Domestic Product (GDP), contributing roughly 45.19 percent in 2020

and 45.93% in 2021. The industrial sector of Ghana is the country's second

largest economic area, contributing roughly 31.6 percent in 2020 and 28.26

percent in 2021 to Ghana's Gross Domestic Product respectively. In Ghana,

the industrial sector is the second largest consumer of petroleum products,

accounting for 9.1 percent of total petroleum product consumption in 2000 and

rising to 466.4 Ktoe in 2019, accounting for 12.1 percent (Energy

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Commission, 2021). Therefore, it is crucial to consider how the industrial

sector of Ghana is affected by volatility of crude oil prices.

Background to the Study

Recent years have seen a resurgence of interest in the investigation of

the effects of changes in oil prices on commercial performance, in part due to

the rise and subsequent fall in crude oil prices that commenced in 2001,

followed by another in 2008, after the subprime mortgage crisis. European

prices for a barrel of Brent crude oil have fallen from $111 in July 2014 to less

than $50 in January 2015. Recent events, like the Covid-19 outbreak, have

caused oil prices to rise or fall throughout the world (Prabheesh et al., 2020).

Supply disruptions that are detrimental to the economy and a decrease in oil

usage came from social isolation measures, company closures, and personnel

cuts in the oil industry caused by infection and quarantine restrictions

(Prabheesh et al., 2020). Boateng et al. (2021) stated that the shocks from

Covid-19 occurred at the same time as the failure agreement among OPEC and

its partners to restrict world oil output. As a result of these factors, there was a

change in the price of crude oil throughout the first and second quarters of

2020. As seen in Figure 1, the price of a barrel of Brent crude fell from $63.60

in January 2020 to $23.34 in April 2020.

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 1: The Trend in Average Monthly Brent Crude Oil Prices.

Source: Bank of Ghana (2020)

According to Jiménez-Rodrguez and Sánchez (2005), the influence of

fluctuating crude oil prices on economic development is relayed through

supply and demand channels. This is the case even when supply and demand

channels are not directly involved. Shocks to the price of oil can either raise or

decrease production expenditures, which can lead to a drop in supply-side

output quality. On the demand side, Jimenez-Rodriguez et al. (2005) claim

that oil price shocks influence consumption of goods and services indirectly

through their effect on disposable income. That is, a rise in the price of oil

lowers consumers' disposable income, which in turn causes them to delay their

purchases. It is possible that the pace of economic activity may decrease as a

result of this condition. Furthermore, oil price shocks may heighten

uncertainty about the economy's future prospects, prompting people to save as

a cautious move, negatively impacting economic activity (Jimenez-Rodriguez

et al., 2005). Because of this, the impact on economic growth will also be

contingent on whether or not the nation is a net exporter or importer of crude

Digitized by Sam Jonah Library

University of Cape Coast [Link]

oil. A surge in crude oil prices oil is good news for economies that rely on

exports, and vice versa. (Jiranyakul, 2016).

The heterogeneous market theory (Müller et al., 1997) and the adaptive

market hypothesis (Asafo-Adjei, Adam & Darkwa, 2021; Boateng et al., 2021;

Li, Huang & Failler, 2022, etc.) suggest that the crude oil market is diverse

and adaptable (Lo, 2004). In the opinion of Müller et al. (1997), players in the

market are irrational, which might change market dynamics over short,

medium, and long investment horizons. Lo (2004) argues that shifts in the

dynamics of the market cause structural shifts in both time and frequency, and

that this correlates to the adaptability of the market. It is essential, for this

reason, to analyze the degree of quantile asymmetric nexus, time-varying co-

movements and conditional causality associated with alterations in the value

of crude oil.

The expansion of the economy of Ghana is susceptible to being

significantly influenced by shifts in both the cost of crude oil and its overall

availability in the country (Centre for Study of Africa Economies, 2014).

According to Awunyo-Vitor et al. (2018), the Ghanaian economy is heavily

dependent on imported crude oil and this demonstrates the crucial role that

crude oil plays in the economy as a result of rising consumption. This is

because Ghana is unable to produce enough of its own crude oil to meet the

country’s demand for petroleum products and electricity generation (Cantah &

Asmah, 2015; Awunyo-Vitor et al., 2018). The Energy Commission (2021)

argues that, despite recent finds, Ghana stays a crude oil importer. Ghana has

generally sought to sell higher-grade oil while importing crude oil for

economic reasons. In the year 2020, a total of 4,843 kilotonnes of crude oil

Digitized by Sam Jonah Library

University of Cape Coast [Link]

were imported, the refineries utilized 461 kilotonnes for manufacturing

petroleum products, while the remaining 382 kilotonnes were utilized to

generate power. as shown in Figure 2.

10,000
9,000
8,000
7,000
Kilotonnes

6,000
5,000
4,000
3,000
2,000
1,000
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Years

Refinery Use Electricity Generation

Figure 2: The Trend in Crude Oil Import in Ghana

Source: Energy Commission (2021)

Ghana's industrial sector is regarded as second largest contributor to

the GDP of the country. Data for 2019 from the Ghana Statistical Service,

(2020) revealed a growth rate of 6.4 percent, compared to 10.6 percent in

2018. The highest growth rate for 2019 was 7.6 percent in the service sector,

followed by 6.4 percent in industry and 4.6 percent in agriculture. The service

sector continues to be the most important sector. Its GDP share climbed from

47 percent in 2018 to 48.2 percent in 2019. The sectors GDP growth rate, on

the other hand, increased from 2.7 percent in 2018 to 7.6 percent in 2019

(Ghana Statistical Service, 2020) as shown in Figure 3.

Digitized by Sam Jonah Library

University of Cape Coast [Link]

18
16
14

Growth Rate of GDP

12
10
8
6
4
2
0
-2 2014 2015 2016 2017 2018 2019 2020
-4
Years

AGRICULTURE INDUSTRY SERVICES

Figure 3: Growth rate of GDP for Agriculture Sector, Industrial Sector and
Services Sector.
Source: Ghana Statistical Service (2020).

The industrial sector, with a 29.88 percent GDP share in 2020, saw its

growth rate decrease from 6.4 percent in 2019 to a negative growth rate of 2.5

percent in 2020 (Ghana Statistical Service, 2020). The industrial sector of

Ghana experienced a negative growth rate in 2020, primarily due to the

disruptive repercussions of the COVID-19 epidemic. Lockdowns, supply

chain disruptions, reduced demand, global economic slowdown, and financial

constraints contributed to the contraction, affecting industries reliant on both

domestic and global markets. Additionally, the decline in oil prices and labor

force challenges further compounded the sector's challenges (Ghana Statistical

Service, 2021; World Bank, 2020). The sector is made up of five subsectors,

namely; manufacturing, mining and quarrying, construction, water and

sewerage, and electricity. The Energy Information Administration (2021),

asserts fuel is one of the primary sources of energy that is utilized in the

commercial and manufacturing sectors. In the manufacturing industry, crude

Digitized by Sam Jonah Library

University of Cape Coast [Link]

oil and products derived from petroleum are put to use as a means of energy,

as feedstock or as unprocessed goods in the production of a wide variety of

intermediate and end-user goods, including plastics, polyurethane, and

solvents. Refineries, for example, use crude oil to make gasoline, kerosene,

gas oil, and other fuel oils, among others (Energy Information Administration,

2021). Because crude oil is a main source of energy for the industrial sector,

fluctuations in its price may have an effect on the percentage of GDP that the

industrial sector contributes.

As a result, it is essential to investigate the impact that the fluctuation

in the price of crude oil has on the industrial sector. The current study

specifically explores their asymmetry, causality, as well as time and frequency

nexus in order to support optimal resource management decisions.

Asymmetric analysis is used to understand how different economic variables

or factors affect a particular sector, such as the industrial sector in this case,

during different phases of their fluctuations. In the context of the effect of

crude oil price fluctuations on the industrial sector of Ghana, asymmetric

analysis is employed to examine whether the sector responds differently to

upward price swings compared to downward movements in crude oil prices.

This approach is informed by the recognition that economic agents

often react differently to both beneficial and detrimental shifts in key factors.

For example, during periods of rising crude oil prices, industries may face

higher production costs, which could lead to decreased profitability and

reduced output. On the other hand, when oil prices decline, industries might

experience cost savings, potentially resulting in increased production and

economic growth. This uneven response to positive and negative changes is

Digitized by Sam Jonah Library

University of Cape Coast [Link]

known as asymmetric behavior. This is done in consideration of the shifts that

occur in both crude oil prices and the industrial sector. The asymmetric

relationship will be answered through the quantile regression approaches as

adopted by prior literature (Adebayo, Rjoub, Akinsola & Oladipupo, 2022;

Archer, Owusu Junior, Adam, Asafo-Adjei & Baffoe, 2022; Barson, Owusu

Junior, Adam & Asafo-Adjei, 2022; Demir, Pesqué-Cela, Altunbas &

Murinde, 2022) to divulge the economy's conditions in terms of stress, normal

and boom. On the other hand, the bi-wavelet approach which examines co-

movements and lead-lag nexus between time series data across time and

frequency would be employed (Armah, Amewu & Bossman, 2022; Asafo-

Adjei, Adam & Darkwa, 2021; Singh, Bansal & Bhardwaj, 2022).

Statement of the Problem

Energy is a crucial factor in the national economic development and

sustainable progress of any nation, including Ghana (Awunyo-Vitor et al.,

2018; Mukhtarov, Humbatova, Mammadli & Hajiyev, 2021). A growing

number of countries throughout the world are becoming more concerned about

the asymmetric, causality and time-frequency varying effect of unstable oil

prices on economic growth (Asafo-Adjei, Adam & Darkwa, 2021; Boateng,

Adam & Owusu Junior, 2021; Ftiti et al., 2016; Mo et al., 2019). The rising oil

prices have been a major challenge for both local businesses and the global

economy (Cantah & Asmah, 2015; Boateng et al., 2021).

In Ghana, consumers of petroleum products including firms bear the

full cost of petroleum use, and the increased prices have led to significant

problems for businesses. These problems include increased transportation and

production costs, reduced profitability, and the need for cost-cutting measures

Digitized by Sam Jonah Library

University of Cape Coast [Link]

such as layoffs and reduced production. The downward review of the Special

Petroleum Tax (SPT) incorporated into Ghana's Petroleum Price Build-Up

(PBU) and the revision and neutralization of the Price Stabilization and

Recovery Levy (PSRL) has an impact on rising petroleum prices for

consumers in Ghana (Tsatsu, 2022). This was intended to lessen the impact of

rising crude oil prices. However, the downward review was unable to stop the

price of petrol from rising by more than 38% between January 2017 and June

2019 to sell at Ghs5.25 per litre. Also, from June 2019 to March 2022, fuel

costs would rise by 64.15%, leading to an increase in price from GHS 5.25 to

GHS 8.618.

This is extremely concerning and detrimental to all economic actors,

notably the industrial sector, whose operations substantially depended on fuel

consumption. The Chief Executive Officer (CEO) of the Association of Ghana

Industries (AGI), Seth Twum Akwaboah, described the business environment

as challenging in the country. In his view, utility tariffs have increased, fuel

prices have increased, the policy rate has increased – leading to high interest

rates, high levels of inflation, depreciation of the cedi, and a variety of

businesses have been forced to cut production (The Business and Financial

Times, 2022).

Statistics on petroleum product consumption by sectors from the

Energy Commission (2021) as shown in Figure 4 revealed that the

transportation sector utilized 1,148.2 Ktoe of petroleum products in 2000,

accounting for 79.9 percent of total petroleum products consumed in the

economy. In 2019, this grew to 2,950.7 Ktoe, accounting for 77.2 percent of

total petroleum product consumption. The industrial sector is the second

Digitized by Sam Jonah Library

University of Cape Coast [Link]

largest user of petroleum products, accounting for 9.1% of total consumption

in 2000 and rising to 466.4 Ktoe, representing 12.1 percent (Energy

Commission, 2021). Hence, the rise in consumption of petroleum products in

light of the corresponding increase in crude oil price would exacerbate the

impact of price volatility on industrial sector output. Accordingly, with this

level of consumption, the industrial sector output at various economic

situations may be vulnerable to crude oil price volatility.

9000
Petroleum Product Consumption by Sector

8000
7000
6000
5000
(Ktoe)

4000
3000
2000
1000
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Years
Residential Industrial Commerce & Service
Agric & Fisheries Transport Non Energy use
Total Consumption

Figure 4: The Trend in Petroleum Product Consumption by Sectors (Ktoe)

Source: Energy Commission (2021)

Policies by the Government of Ghana to cut subsidies on petroleum

products could cause fluctuations in petroleum products prices (Adjei-Mantey

& Takeuchi, 2023; Kojima, 2016). This can further lead to adverse influence

on the output of the industrial sector asymmetrically depending on the size and

growth of the industries. The reduction in subsidies on petroleum products in

periods such as 2013 and 2015 led to an eventual increase in production cost

10

Digitized by Sam Jonah Library

University of Cape Coast [Link]

of firms which dwindled output. For example, the government of Ghana in

2015 set aside US$12.5m for subsides, compared with the US$150m in 2014

(Economist Intelligence, 2015).

Many theoretical and empirical studies have been conducted on the

topic of the relationship between fluctuations in oil prices and economic

growth. These studies used a variety of perspectives and methods, but they

mostly focused on developed economies (Burakov, 2016; Koirala and Ma,

2020; Humbatova, 2019; Le & Chang, 2015; Mo, Chen, Nie & Jiang, 2019;

Yazdani & Noor 2015). However, there are few references to Ghana and other

developing countries (Ogunsakin & Oloruntuyi, 2017; Cantah & Asmah,

2015; Dogah, 2015; Donkor, 2018; Lin, Wesseh, & Appiah, 2014).

The vast majority of these studies provide evidence that supports a

negative link between fluctuations in the price of crude oil and economic

growth, however some of these studies also challenged the validity of the

negative association (Mgbame & Donwa, 2015). The few studies on crude oil

price dynamics conducted in developing countries have not focus the

discussions on the industrial sector (Asafo-Adjei, Adam & Darkwa, 2021;

Boateng et al., 2021; Boateng et al., 2022; Dramani & Frimpong, 2020;

Kassem, Khoiry & Hamzah, 2019). For instance, Asafo-Adjei, Adam and

Darkwa (2021) and Boateng et al. (2021) explored the impact of fluctuations

of crude oil price on stock returns in Africa but did not focus on the industrial

sector. In addition, Boateng et al. (2022) investigated the influence of crude oil

price on the real sector of Ghana.

Nonetheless, studies conducted on the impact of crude oil price

volatility on industrial sector output (Abdulkareem & Abdulkareem, 2016;

11

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Ahmed et al., 2017; Aimer, 2016; Akalpler et al., 2018; Al-Risheq, 2016; Al-

Sasi et al., 2017; Awunyo-Vitor et al., 2018; Eksi et al., 2011; Hau, Zhu,

Huang, and Ma, 2020; Iganiga et al., 2021; Ogunsakin et al., 2017; Riaz et al.,

2016; Yu et al., 2022; Zhang et al., 2022) employ methodologies that do not

account for diverse economic conditions of the industrial sector. In this regard,

prior studies have not yet addressed the quantile asymmetric and conditional

causality effects of crude oil price volatility on diverse economic conditions of

industrial sector output.

Also, considering the degree of heterogeneity and adaptability of the

crude oil market in relation to economic growth, prior studies that have used

the industrial sector output as proxy have failed to assess the time-frequency

nexus between the variables in a developing country context. The inclusion

of a time-frequency nexus analysis in this study holds significant policy

implications for both the industry and public policy debate. The time-

frequency analysis allows for a detailed examination of how the relationship

between crude oil price volatility and industrial sector output evolves over

different time scales. This information is crucial for policymakers, industry

leaders, and stakeholders in making informed decisions. For instance,

understanding how the impact of crude oil price fluctuations varies across

short-term and long-term periods can guide the development of targeted

policies that address immediate shocks as well as long-term sustainability.

Moreover, this nuanced analysis can aid in the formulation of adaptive

strategies to mitigate the adverse effects of oil price volatility on the industrial

sector, enhancing the resilience of the sector in the face of changing market

conditions.

12

Digitized by Sam Jonah Library

University of Cape Coast [Link]

The unique contribution of this study lies in its focus on the developing

country context of Ghana and the specific examination of the industrial

sector's response to crude oil price volatility using a time-frequency nexus

analysis. While prior studies have often used the industrial sector's output as a

proxy without considering the intricate temporal dynamics, this study delves

into the time-frequency relationship between crude oil price fluctuations and

sectoral output. This approach acknowledges the heterogeneity and

adaptability of the crude oil market and its interaction with the industrial

sector in a developing economy, offering insights that can guide policy

decisions tailored to Ghana's unique circumstances. Furthermore, the study

contributes to the broader global understanding by examining these dynamics

in the context of a developing country, potentially providing valuable insights

for other economies facing similar challenges and opportunities. The emphasis

of this study compliments these studies by looking at the effect of crude oil

price volatility on the output of the industrial sector in Ghana. This sector is

the nation’s second largest user of refined crude oil products and it is

important to evaluate its vulnerability to crude oil price fluctuations.

Empirical analysis is performed on the vulnerabilities of the

aggregated and the disaggregated stages of the industrial sector output to crude

oil price volatility. Comparison can therefore be made between the outcomes

of the aggregated and disaggregated levels of industrial sector output. This is

needed to decipher the extent to which the sub-sectors reflect the overall sector

for policy decisions in the management of resources in the country. This is

relevant because, oil price volatility does not affect all sectors equally

(Taghizadeh-Hesary, Rasolinezhad, & Kobayashi, 2015). That is to say,

13

Digitized by Sam Jonah Library

University of Cape Coast [Link]

certain business sectors are more severely affected than others leading to

asymmetric and heterogeneous nexus. The subsectors of the industrial sector,

including manufacturing, construction, mining and quarrying, water and

sewerage, and electricity, are likely to be affected differently by crude oil price

volatility due to their inherent characteristics and dependencies. The

manufacturing sector is intricately tied to crude oil prices as it relies on

petroleum products both as raw materials and energy sources. Fluctuations in

oil prices can significantly impact production costs, particularly for industries

dependent on plastics, chemicals, and transportation-related products. During

periods of high oil prices, manufacturing industries could experience cost

escalation, potentially leading to reduced profitability and increased consumer

prices. Conversely, lower oil prices might offer cost relief for these sectors,

potentially stimulating production and economic growth. Construction

activities are closely linked to the costs of transportation, machinery, and

energy, which can be heavily influenced by oil prices. Higher oil prices can

lead to increased costs for construction materials and equipment, potentially

causing delays and cost overruns in projects. Lower oil prices, however, could

mitigate some of these cost pressures, potentially benefiting the construction

industry.

The mining and quarrying sector can be affected by oil price volatility

through its impact on transportation costs, as well as energy-intensive mining

processes. Increased oil prices could drive up operational costs, affecting

profit margins in mining and quarrying activities. On the other hand, lower oil

prices might ease cost pressures, allowing the sector to maintain or even

expand its operations. While Water and Sewerage sector is not as directly tied

14

Digitized by Sam Jonah Library

University of Cape Coast [Link]

to oil prices as others, it does have some energy and operational costs related

to transportation and equipment maintenance. Depending on the extent of

these dependencies, fluctuations in oil prices could influence the sector's

operational expenses. The electricity sector's operations are affected by both

the cost of fuel used for power generation and the broader energy market

dynamics. Higher oil prices can lead to increased fuel costs for thermal power

plants, potentially impacting electricity prices and supply reliability.

Conversely, lower oil prices might provide cost relief for the energy sector,

which could translate into more stable electricity prices and availability.

In summary, the diverse dependencies of these subsectors on

petroleum products, energy, and transportation costs create varying degrees of

vulnerability to crude oil price volatility. This lies in the fact that each

subsector's operations and cost structures interact differently with changing oil

prices, leading to asymmetric impacts. A comprehensive analysis of these

subsectors will provide valuable insights for policy decisions and resource

management strategies, ensuring a targeted and effective approach to

addressing the heterogeneous effects of oil price fluctuations across Ghana's

industrial landscape. The paper addresses these issues in this context, the

quantile asymmetric, causality and time- and frequency-dependent impact of

crude oil price volatility on industrial sector output. It is against this backdrop

that the study investigates the quantile asymmetric, causality and time- and

frequency-dependent impact of crude oil price volatility on industrial sector

output.

15

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Purpose of the Study

The purpose of this study is to investigate the effect of crude oil price

volatility on industrial sector output in Ghana.

Research Objectives

The specific objectives include:

1. Estimate the asymmetric relationship between realised crude oil price

volatility and industrial sector output.

2. Determine the conditional causality of realised crude oil price volatility

to industrial sector output.

3. Analyse the responsiveness of the industrial sector output to realised

crude oil price volatility across time and frequency.

Research Questions

The study is guided by the following research questions;

1. What is the asymmetric relationship between realised crude oil price

volatility and industrial sector output?

2. What is the conditional causality of realised crude oil price volatility to

industrial sector output?

3. What is the responsiveness of the industrial sector output to realised

crude oil price volatility across time and frequency?

Significance of the Study

This research attempts to examine the effect that variations in the price

of crude oil have on the production of the industrial sector. Ghana, as a nation

heavily dependent on oil imports for both its industrial processes and energy

16

Digitized by Sam Jonah Library

University of Cape Coast [Link]

requirements, remains exposed to the variations of global crude oil price

fluctuations. The findings of this study hold substantial significance, offering

valuable insights into the extent to which crude oil price volatility influences

industrial output within the country. This empirical knowledge can serve as a

vital resource for policymakers and industry stakeholders, enabling them to

make informed decisions concerning resource allocation, investment

strategies, and effective risk management practices in a dynamic economic

environment.

Delimitations

The purpose of the study was to investigate how the output of Ghana's

industrial sector was affected by the fluctuation in the price of crude oil, which

in this case, realised crude oil price volatility. Five sub-sectors in the industrial

sector of Ghana were utilised. These were mining and quarrying,

manufacturing, electricity, water and sewerage, construction. The industry

sector average was also included in this study for further inferences. In the

study, researchers utilised a monthly time series data set that covered the

period starting in January 2001 and ending in December 2020. This area was

selected due to data accessibility. The study employed the quantile regression

bi-wavelet, and conditional causality in quantile approaches.

Limitations

The research did not concentrate on the whole economy. As crude oil

is thought to be one of the key sources of energy, it is possible that it will have

an effect on other parts of the economy. This study focuses solely on the

industrial sector. Another drawback is the lack of monthly data on industrial

sector output; the only information available was on a quarterly basis. In

17

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Eviews-10, the method developed by Chow and Lin (1971) was applied in

order to produce the monthly series. Because there are no restrictions on the

sub-periods that can be used when disaggregating data with this approach,

there are no issues to the reliability of the findings when employing it.

Also, the quantile regression technique used in this study only

considers the quantiles of the industrial sector output as the dependent

variable. Hence, the study is not able to reveal quantile-on-quantile effects of

both dependent and independent variables. However, application of the

quantile regression approach is ideal in controlling for other macroeconomic

factors relative to the quantile-on-quantile technique which is limited to only

two variables. Furthermore, despite the fact that the conditional causality in

quantile approach examines causality between two variables, it is able to

investigate causality for relationships that are non-linear. The bi-wavelet on

the other hand shows the interconnectedness between two variables without

controlling for a third variable, as in the case of partial wavelet approach.

However, it is a useful tool which examines the interconnectedness of two

variables across time and frequency indicating leading and lagging variables.

Organisation of the Study

There are five chapters in the research. The study's introduction,

problem statement, purpose, goals, and questions are all covered in the first

chapter along with its importance, constraints, and organizational structure.

The second chapter summarizes empirical findings from earlier studies that are

relevant to this research and evaluates the literature on crude oil prices and

their impacts on different economic sectors. The methodology of the study, the

research design, the research technique, and data processing and analysis are

18

Digitized by Sam Jonah Library

University of Cape Coast [Link]

all outlined in Chapter Three. The findings of the empirical investigation are

explained in Chapter Four. The last chapter presents the study's findings,

conclusions, and suggestions.

CHAPTER TWO

LITERATURE REVIEW

Introduction

This chapter covers reviews on the effect of crude oil price volatility

on the industrial sector's output. The study's underlying ideas, a conceptual

framework for oil price volatility, industrial output, and a review of pertinent

empirical research are explored in this chapter.

Theoretical Review

The theoretical linkages among the variables are presented to highlight

some relevant hypotheses such as heterogenous and adaptive market

hypotheses. the theoretical linkages among the variables are presented to

highlight some relevant hypotheses such as heterogenous and adaptive market

hypotheses. As a first illustration, the symmetric/linear association growth

theory is presented. Next, the theoretical linkages among the variables are

presented to highlight some relevant hypotheses such as heterogenous and

adaptive market hypotheses.

The symmetric/linear relationship growth theory

Proponents of the symmetric/linear connection theory of growth

(Hamilton, 1983; Hooker, 1986) argue that fluctuations in the growth rate of

19

Digitized by Sam Jonah Library

University of Cape Coast [Link]

production are caused by fluctuations in oil prices (Taofik, 2018). Their

arguments were based on what occurred in the oil market and how it affected

the economies of oil exporting and importing nations between 1948 and 1972.

Hooker (2002) conducted extensive empirical research between 1948 and

1972 and found that oil price changes significantly influenced production

growth. According to Taofik (2018), Laser (1987) supports the notion that

there is a symmetric link between the volatility of oil prices and economic

growth. A rise in oil prices would inevitably lead to a drop in GDP, whereas

the impact of a drop in oil prices on GDP is equivocal due to the fact that its

influence differs from nation to nation (Taofik, 2018).

According to Eagle (2017), there are few inputs that have both

symmetrical and asymmetric effects on the performance of the macroeconomy

including oil. This means that changes in the price of oil can cause

unanticipated repercussions on economic growth. He was of the belief

alterations in the price of oil stifle growth through a variety of routes, such as

increases in production costs and anticipation of higher inflation. In addition,

he asserted that a rise in the price of oil can result in improvements to a

nation's transportation and production levels as well as the stability of its

financial system. It is against this backdrop that this study examines the

asymmetric nexus between crude oil volatility and performance of the

industrial sector.

Theoretical channels of the influence of crude oil price volatility on

industrial output

According to Bugshan (2021), volatility in oil prices affects industrial

output through three pathways, namely: output level, monetary policy, and

20

Digitized by Sam Jonah Library

University of Cape Coast [Link]

market value. The consequences of an increase or reduction in the price of oil

for the trajectory of output level are different for each economy. These

implications might be positive or negative. An increase in the price of oil has a

negative impact on the output of nations that import oil since it leads to greater

production costs and reduces the buying power of consumers, both of which

work to lower output. This connection is supported by the findings of

empirical studies conducted in industrialized markets that are net importers of

oil (Bilgin, Gozgor & Karabulut, 2015; Tang, Wu & Zhang, 2010). A

plausible explanation for this could be that increased volatility makes

forecasting future economic growth and demand in markets more difficult; as

a result, businesses postpone investment (Henriques & Sadorsky, 2011).

As indicated, monetary policies are another pathway through which oil

price volatility (OPV) affects industrial output (Bugshan, 2021). The increased

cost of production caused by a rise in the price of oil will be passed on to

customers in the form of higher prices. As a direct consequence of this, the

rate of inflation in the nation will accelerate. As a standard response, the

Central Bank will boost interest rates in order to combat the inflationary

pressure. This will have two ramifications. First, the discount rate for expected

future cash flows will increase (Cologni & Manera, 2008). Also, the higher

cost of corporate financing will cause project investment to be delayed

(Henriques & Sadorsky, 2011; Phan, Tran & Nguyen, 2019).

According to Basher and Sadorsky (2006) and Bugshan (2021), the

nature of an industry's operations determines whether OPV will have a

favourable or adverse impact on future cash flow. Profits and stock prices for

oil companies would benefit from a rise in the price of oil. Oil and gas

21

Digitized by Sam Jonah Library

University of Cape Coast [Link]

companies gain from increasing oil prices because their profit margins expand.

Yet, sectors dependent on oil for their production are expected to see a decline

in profits. This is because of the direct correlation between rising oil prices

and higher production costs (Bugshan, 2021).

Another potential mechanism is through changes in demand. Volatility

in crude oil prices can impact consumer and business confidence, leading to

changes in consumption patterns and investment decisions (Henriques &

Sadorsky, 2011). For instance, when crude oil prices increase, consumers may

reduce their demand for goods and services that require high energy use,

leading to a decline in demand for related products and services (Sorrell,

2015). Moreover, businesses may delay or cancel investment decisions in new

projects or equipment that require high energy consumption (Ang, 1992). This

could lead to a decline in economic growth in the industrial sector.

Accordingly, the impact of the fluctuations in crude oil prices on

industrial output becomes heterogeneous and adaptive across time and

frequency due to the behavioral intentions of investors which would alter in

response to the changes in the market’s dynamics (Cornell, 2018; Owusu

Junior et al., 2021). The heterogeneous market hypothesis (Müller et al., 1997)

and the adaptive market hypothesis (Boateng et al., 2021; Li, Huang & Failler,

2022) suggest that the crude oil market is diverse and adaptable. Therefore, the

effect of crude oil price on economic growth might not be the same across

economic conditions of stress, normal and boom as well as time and

frequency.

Müller et al. (1997) argue that market participants are irrational, which

might change market dynamics over short, medium, and long investment

22

Digitized by Sam Jonah Library

University of Cape Coast [Link]

horizons. Lo (2004) argues that shifts in the dynamics of the market cause

structural shifts in both time and frequency, and that this correlates to the

adaptability of the market. Hence, the markets become nonlinear, asymmetric

and nonstationary (Owusu Junior et al., 2021). This demands empirical

methods that can decipher the nexus between fluctuations in crude oil and

industrial output. In this regard, the current study explores the asymmetric as

well as time and frequency co-movements between crude oil price volatility

and industrial output using the quantile regression and bi-wavelet approaches

respectively.

Conceptual Review

Oil price volatility

Volatility in prices is a common economic process that occurs almost

every day within any given financial year. According to Lin et al. (2013), the

term "price volatility" describes the degree to which prices go up or down over

a period of time. In a well-functioning market, prices are indicative of present

and forecasted supply and demand situations, as well as variables that may

have a substantial impact on them. In situations where market prices change

often within extremely short time intervals, then it can be concluded that there

is high price volatility for that particular product (Chen and Hsu, 2012). Thus,

an economy is described as having low price volatility when market prices are

fairly stable for longer periods. There can be volatility in the prices of all

economic goods, including gold, timber, cocoa, coffee, and cashew.

According to the findings of Ogiri, Amadi, Uddin, and Dubon (2013),

volatility is the measure of the propensity of oil prices to rise or decrease

abruptly within a period of time, such as a day, a month, or a year. This

23

Digitized by Sam Jonah Library

University of Cape Coast [Link]

tendency may be seen in the context of crude oil price volatility. According to

Lee (1998), which is referenced in Mgbame et al. (2015), volatility is defined

as the standard deviation in a given time, and volatility has an instantaneous

negative and big influence on economic development. In a nutshell, volatility

may be defined as the degree to which the price of a commodity fluctuates

over time (increase and fall). Oil price volatility can have huge repercussions

for businesses and the global economy. (Council on Foreign Relations, 2016).

Historically, the international commerce of crude oil has been crucial

to the development of the global economies (Taofik, 2018). One of the few

production inputs that may potentially create a recession is oil (Hamilton,

2009; Tverberg, 2012). Oil price volatility stifles growth via a variety of

mechanisms, ranging from increased production costs to inflation expectations

(González and Nabiyev, 2009). Oil price shocks over the past few decades

have dampened global economic expansion (Tang, Wu, & Zhang, 2010). As

oil supplies were interrupted by a succession of political events in the Middle

East in the 1970s and 1980s, prices dropped throughout the world (Tang et al.,

2010). According to Awunyo-Vitor, Samanhyia, and Bonney (2018), the

decades of the 1970s and 1980s made it abundantly evident how reliant

developed economies are on crude oil.

Major economies like the United States and the United Kingdom have

seen slowing growth, rising unemployment, and rising prices across board.

Several industrialized economies went through the stagflationary period

(Awunyo-Vitor et al., 2018). After the initial interruption, several more events

have occurred, each of which has reduced crude oil output and increased

prices. These incidents include the Iraq-Iran War in 1978, the Gulf War in

24

Digitized by Sam Jonah Library

University of Cape Coast [Link]

1990, the Asian Economic Crisis in 1977, the Iraq War in 2003 (Tunyo,

Armah, Cantah, & Suleman, 2021), and the COVID-19 Pandemic (Boateng et

al., 2021; Boateng et al., 2022; Prabheesh, Padhan, & Garg, 2020). Therefore,

more evidence-based research must understand the effects such volatilities can

have on industrial sector output in the long-run and short-run.

Industrial sector output

Ghana's industrial sector has experienced significant growth and

transformation over the past few decades (Ackah, Adjasi & Turkson, 2016;

O'Neill, 2022; Tunyo et al., 2021). This can be seen in the sector’s

contribution to GDP, own-growth, sub-sectors, ability to offer employment,

coupled with the challenges thereof. The industrial sector in Ghana comprises

several sub-sectors, including manufacturing, construction, water and

sewerage, electricity, and mining and quarrying (Ghana Statistical Service,

2020). The manufacturing sector is the largest sub-sector and includes food

and beverages, textiles and garments, and pharmaceuticals. Figure 5 shows the

growth rates of the sub-sectors of the industrial sector.

35
30
25
20
Growth Rates (%)

15
10
5
0
-5 2014 2015 2016 2017 2018 2019 2020
-10
-15
Years

Mining and Quarrying Manufacturing Electricity

Water and Sewerage Construction

25

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 5: Growth Rate of the Sub-sectors of the Industrial Sector

Source: Ghana Statistical Service (2020)

From Figure 5, the manufacturing sector was the least performing sub-

sector in 2014 followed by the construction sector. During this period, water

and sewerage sector was the highest performing, followed closely by mining

and quarrying. However, in 2015 and 2020, the growth rates of the mining and

quarrying sector were the lowest despite its maximum growth from 2017 to

2019. Almost all the sub-sectors record a negative growth rate at particular

years suggesting that the output of the industrial sector has experienced rapid

variations over time.

The industrial sector faces several challenges, including inadequate

infrastructure, limited access to finance, and low levels of technology and

innovation (Ackah, Adjasi & Turkson, 2016; Ghana Statistical Service, 2020;

Tunyo et al., 2021). These challenges have hindered the sector's ability to

compete internationally. In conclusion, while Ghana's industrial sector has

shown promising growth, there is still room for improvement to fully realize

its potential and address the challenges it faces.

Relationship between industrial sector output and crude oil price

volatility

Crude oil is a key input cost for many industries, particularly those that

rely on transportation and manufacturing (Al-Risheq, 2016). When the price of

crude oil rises, it can increase the cost of inputs for these industries, which can

reduce profitability and potentially lead to lower production levels (Boateng et

al., 2021). On the other hand, the price of crude oil can also impact consumer

spending patterns, as higher oil prices can lead to higher prices for goods and

services across the economy (Phan, Tran, Nguyen & Le, 2020). If consumers

26

Digitized by Sam Jonah Library

University of Cape Coast [Link]

are forced to spend more on essentials like gasoline and heating oil, they may

have less disposable income available to spend on other goods and services,

which can in turn impact demand for industrial goods (Okwu, Akpa, Oseni, &

Obiakor, 2020).

The price of crude oil is also closely tied to macroeconomic factors

like inflation, interest rates, and global economic growth (Asafo-Adjei et al.,

2020; Asafo-Adjei, Boateng et al., 2021; Boateng et al., 2022; Kasongwa, &

Minja, 2022). For example, if the global economy is experiencing a period of

rapid growth, demand for crude oil may increase, which can drive up prices.

Conversely, if interest rates rise, it can lead to a decrease in demand for oil and

other commodities, which can lead to lower prices.

Additionally, geopolitical events like conflicts, natural disasters, and

trade disputes can also impact the price of crude oil (Gong, Feng, Liu &

Xiong, 2023; Zhang, Wang, Xiao & Zhang, 2023). For example, if there is a

disruption in oil supply due to a natural disaster or geopolitical conflict, it can

lead to a sudden spike in prices that can impact the industrial sector. In

summary, the relationship between crude oil price volatility and industrial

sector output is complex, with many different factors at play. The price of

crude oil can impact input costs, consumer spending, macroeconomic factors,

and geopolitical factors, all of which can in turn impact the output of industrial

sectors in various ways.

Conceptual Framework

This section shows the study’s conceptual framework. The sustainable

progress of any country is built on a number of elements, one of which is

energy, and Ghana is no exception. In addition to a number of other important

27

Digitized by Sam Jonah Library

University of Cape Coast [Link]

aspects, crude oil is one of the most important components in the production

process of the global economy. Crude oil price volatility describes the degree

to which prices go up or down over a period of time. Crude oil price volatility

has the tendency of affecting output of industrial sector as it serves as one of

the inputs of industrial production.

Based on literature, some other macro-economic variables which could

affect industrial output were included in the framework as control variables

(Abokyi et al., 2018; Al-Risheq, 2016; Iganiga, Anyanwu, Ikubor, & Ojima,

2021). They include energy consumption, foreign direct investment, and

interest rate. Industrial output may also be affected by the energy consumed as

energy serves as one major input of industrial production. According to

Abokyi et al. (2018), the unpredictable electricity power supply is regarded as

the number one issue affecting the industrial sector in Ghana. Because of this,

energy consumption is incorporated into the model as a control variable.

Interest rates also have the tendency to affect industrial output. Butler

(2022) opined that lower interest rate spurs growth while higher interest rate

reduces spending and investment, causing a reduction in industrial output.

Most industries in Ghana finance their businesses through loans from banks

and may be vulnerable to interest rates. Interest rate is therefore included in

the model as a control variable. Foreign direct investment (FDI) can be

described as an ownership stake in a foreign company or project made by an

investor, company, or government from another country. Iddrisu, Adam, and

Halidu (2015) looked at the impact of FDI on Ghana's industrial sector's

output from 1980 to 2013. According to the report, foreign direct investment

28

Digitized by Sam Jonah Library

University of Cape Coast [Link]

significantly improves Ghana's industrial sector's output over time. This

justifies the inclusion of FDI as a control variable in the model.

Overall, the conceptual framework posits that fluctuations in crude oil

prices can have a significant impact on the output of the industrial sector.

When crude oil prices are high, the cost of production for the industrial sector

firms increases, leading to a decrease in profitability and output. On the other

hand, when crude oil prices are low, the industrial sector firms can benefit

from lower production costs and increased profitability. The control variables

play a role in mitigating the impact of crude oil price volatility on the

industrial sector. Higher levels of energy consumption can increase the

vulnerability of the industrial sector to crude oil price volatility, whereas

foreign direct investment can provide the sector with additional resources and

funding to mitigate the effects of volatility. Interest rates can also influence the

industrial sector's output by affecting the cost of borrowing for firms. Figure 6

presents a conceptual framework that shows the relationship between crude oil

price volatility and industrial output.

Crude Oil Price Volatility Industrial Output

Energy Consumption

Foreign Direct Investment

Interest Rate

Oil Price
29

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 6: Conceptual Framework

Source: Author’s Construct (2023)

Empirical Review

There is a plethora of research available on the effects of fluctuations

in the price of crude oil. The vast majority of these research focused on the

effect that fluctuating crude oil prices have on economic development,

whereas just a small number of studies looked at the impact of other

macroeconomic factors.

Asymmetric nexus

Abdulkareem and Abdulkareem (2016) conducted a comprehensive

analysis of macroeconomic and oil price volatility within the context of

Nigeria. They employed Generalized Autoregressive Conditional

Heteroskedasticity with Mean (GARCH-M), Exponential Generalized

Autoregressive Conditional Heteroskedasticity (EGARCH), and Threshold

Generalized Autoregressive Conditional Heteroskedasticity (TGARCH), while

considering different time frames, including daily, monthly, and quarterly

data. Their investigation unveiled that all the examined macroeconomic

variables, encompassing real gross domestic product, interest rates, exchange

rates, and oil prices, exhibited pronounced levels of volatility. Notably, the

asymmetric models (TGARCH and EGARCH) outperformed the symmetric

models (GARCH (1 1) and GARCH-M) in their ability to predict volatility

trends. This research underscored the significant impact of oil price

fluctuations on the overall economic instability of Nigeria. Consequently, the

30

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Nigerian economy proved susceptible to both internal factors, such as the

instability of interest rates and real GDP, and external factors, including

fluctuations in exchange rates and oil prices.

Hau, Zhu, Huang, and Ma (2020) conducted an original investigation

into the relationship between the volatility of crude oil prices and China's

agricultural commodity futures. They employed a unique quantile-on-quantile

regression technique to explore the heterogeneous dependence between these

two variables. To estimate the conditional volatility, the researchers utilized a

dynamic model with time-varying parameters and stochastic volatility. Their

findings revealed diverse dependencies between crude oil price volatility and

agricultural futures volatility across various quantiles. Notably, as agricultural

volatility increased in higher quantiles, the spillover of absolute volatility also

grew, indicating a heightened interdependence. This interplay of volatility

exhibited differences in response to market conditions, distinguishing between

turbulent and stable market environments. Furthermore, the study highlighted

those extreme quantiles of oil volatility, whether exceptionally high or low,

exerted a substantial impact, while the behavior of agricultural volatility

remained unaffected during normal oil market conditions. Moreover, the

research unveiled the persistence of volatility dynamics over time,

emphasizing the substantial variations in the influence of volatility on returns.

In 2016, Al-Risheq conducted a comprehensive study examining the

influence of oil prices and various other significant factors on manufacturing

output across 52 developing nations. The study employed a fixed effects

model to analyse data spanning from 1970 to 2012. The study's findings

revealed a noteworthy and adverse relationship between high oil prices and

31

Digitized by Sam Jonah Library

University of Cape Coast [Link]

industrial production. It was evident that elevated oil prices had a substantial
.

detrimental impact on the manufacturing sector. Consequently, the author

concluded that emerging nations, heavily reliant on oil imports, are

particularly vulnerable to negative oil price shocks. Such shocks pose a

significant threat to both industrial production and overall economic growth in

these countries (Al-Risheq,2016).

In addition, Ahmed, Osama Daudpota, and Kashif (2017) used data

from July 2005 through June 2015 using vector auto-regression modeling

approaches to explore the influence of fluctuations in oil prices on production

at the industrial level. Results revealed that high oil price volatility (OPV) was

associated with a significantly higher likelihood of adverse effects on

industrial performance. The study showed that the interaction between OPV

and industrial performance was mediated by the effects of inflation. That is,

whenever there was high volatility in oil prices, it resulted in high inflation

rates, which made it difficult for industries to meet their target output or

productivity level. Hence, supporting the claim that OPV has a considerable

short- to medium-term influence on industrial productivity.

In a similar context, Iganiga, Anyanwu, Ikubor, and Ojima (2021)

conducted an investigation into the relationship between oil price fluctuations

and industrial production levels in Nigeria. The study aimed to uncover both

the symmetric and asymmetric impacts of oil price fluctuations on the

Nigerian industrial sector, utilizing two distinct modeling approaches: the

Autoregressive Distributed Lag (ARDL) model for linear effects and the

Nonlinear Autoregressive Distributed Lag (NARDL) model for nonlinear

effects. According to the short-term linear ARDL model, the research

32

Digitized by Sam Jonah Library

University of Cape Coast [Link]

indicated that higher oil prices tend to positively affect the output of the

building and overall industrial production. Conversely, these elevated oil

prices were found to have a negative influence on the efficiency of the

manufacturing subsector. On the other hand, the long-term nonlinear analysis

revealed that oil price shocks, whether increases or decreases, had diverse and

distinct effects on the industrial sector and its constituent components.

Specifically, a surge in oil prices was associated with a decline in both

aggregate industrial production and the manufacturing index. Conversely, the

data suggested that the output in the building and construction sector

experienced some growth during such price spikes. Moreover, a long-term

decrease in oil prices was linked to an increase in industrial production levels.

Similarly, a group of researchers in Pakistan delved into the

repercussions of oil price volatility (OPV) on the nation's overall

manufacturing output, as outlined by Riaz, Sial, and Nasreen in 2016. In their

study, the authors employed both the EGARCH-in-Mean model and the

ARDL regression model to analyse monthly time series data spanning from

2001 to [Link] results of their investigation unveiled a non-linear

relationship between fluctuations in oil prices and the quantity of

manufactured goods produced in Pakistan. Initially, the country's

manufacturing production exhibited an uptick in response to heightened oil

price volatility. However, the study identified a critical juncture where

manufacturing production began to dwindle even as OPV continued to

increase.

In 2022, Ahmad, Iqbal, Khan, Han, Vega-Muñoz, and Ariza-Montes

conducted a comprehensive analysis examining the macroeconomic

33

Digitized by Sam Jonah Library

University of Cape Coast [Link]

repercussions of crude oil shocks across South Asian countries, including

Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri

Lanka. The study encompassed time series data spanning from 2000 to 2020

and employed two distinct analytical models: the impulse response function

and the vector autoregression model. The research outcomes underscored the

remarkable sensitivity of macroeconomic indicators to even minor fluctuations

in oil prices, emphasizing the significant impact these shifts can exert on the

socio-economic landscape of a region. Notably, the variance decomposition

analysis revealed that each nation in the South Asian region responded

uniquely to crude oil price fluctuations. These diverse responses were

reflective of the specific macroeconomic fundamentals, independent policy

measures, sectoral structures, and inherent national disparities prevalent within

each country (Ahmad et al., 2022).

Conditional causality evidence

Ogunsakin and Oloruntuyi (2017) conducted research in Africa to

investigate the link between OPV and the macroeconomic performance of two

of the continent's top net oil producing countries (Angola and Nigeria). The

research looked at quarterly data covering the years 1990 to 2014 and applied

the E(GARCH), the Granger Causality test and the Structural Vector

Autoregressive (SVAR). The rate of growth of the GDP the currency rate,

foreign interest, and the price of oil in the globe. The variables of interest were

the inflation rate, the domestic real interest rate, and the amount of money in

circulation. According to the findings, the fluctuation in oil prices has a

relatively insignificant effect on the rates of growth of both nations' gross

domestic products (Ogunsakin and Oloruntuyi 2017).

34

Digitized by Sam Jonah Library

University of Cape Coast [Link]

In 2018, Awunyo-Vitor, Samanhyia, and Bonney conducted a study in

Ghana aimed at investigating the direct relationship between oil price

fluctuations and economic development. The research employed various

statistical tests, including the unit root test, the Johansen co-integration test,

and the Granger causality test, to analyze data gathered from 1970 to 2012.

The study's findings revealed an inverse correlation between oil price shifts

and economic growth, suggesting that increases in oil prices tend to be

associated with decreases in economic development. However, when viewed

across a much broader time frame, the influence of fluctuating oil prices on

economic development was found to be statistically insignificant.

Furthermore, the Granger causality test identified a unidirectional cause-and-

effect relationship, indicating that rising oil prices had a causal impact on

expanding economies. In summary, the research indicated that oil price

fluctuations have a minimal impact on the growth of the Ghanaian economy.

Consequently, it underscored the importance of pursuing measures to promote

economic growth that are independent of oil price fluctuations (Awunyo-Vitor

et al., 2018).

In 2011, Eksi, Izgi, and Senturk conducted a study assessing the effects

of crude oil price volatility on industrial production in select Organisation for

Economic Co-operation and Development (OECD) nations. The research drew

its conclusions from the analysis of monthly time series data covering the

period from 1997 to 2008, employing both the Granger causality test and the

Johansen co-integration approach. The study's results unveiled a significant

short-term causal relationship between crude oil prices and industrial

production in all examined countries, except for France. Interestingly, in the

35

Digitized by Sam Jonah Library

University of Cape Coast [Link]

case of France, the causality was observed to run from industrial production to

oil prices in the short term. Furthermore, for the United States, an error

correction mechanism was employed, revealing a long-term causal

relationship from oil prices to industrial production. These findings

collectively suggest that fluctuations in crude oil prices do indeed exert an

impact on the industrial production index in these studied nations.

Akalpler and Bakar (2018) evaluated the influence of variations in oil

prices on economic development between 1981 and [Link] analyze this

relationship, they utilized both the Granger causality test and VECM. Their

findings demonstrated a positive association between the true effective

exchange rate and the price of oil and growth in the economy, but a negative

correlation between government expenditure and inflation. Additionally, they

found that changes in oil prices had a causal effect on economic growth and

the exchange rate, while changes in the exchange rate had a causal effect on

inflation. Through variance decomposition analysis, they observed that OPV

was the largest driver of variation in economic growth and exchange rates,

whereas the exchange rate was the main trigger of variability in inflation,

followed by oil prices.

Al-Sasi, Taylan, and Demirbas (2017) conducted a study on how

―fluctuations in oil prices affect GDP of the United Arab Emirates (UAE). The

aim of the research was to identify in what way oil prices changes influence

macroeconomic indicators in the UAE. They employed several procedures

including Ordinary Least Squares (OLS), ARDL, Augmented Dickey-Fuller

(ADF), and Granger causality methods to analyze the effect of oil prices on

the UAE's economic development from 2001 to 2020. This study revealed

36

Digitized by Sam Jonah Library

University of Cape Coast [Link]

there was a direct correlation between oil price changes and UAE's GDP in

both the immediate and long-term. Consequently, if the essential corrective

efforts are not implemented, a drop in oil prices might represent a long-term

danger to the UAE's economic security.

Time and frequency nexus

Yu, Guo, and Chang (2022) conducted a study ―examining the

relationship between oil price volatility and economic performance during the

COVID-19 and financial crises of 2007-2008. They employed time and

frequency domains simultaneously to look into how shifts in oil prices

affected macroeconomic performance during these crises. By using Wavelet

analysis, the researchers explored the interconnection between oil price shocks

and economic activities during these crises They suggested that during

financial crises, both the economy and oil prices exhibited considerable

strength, but the COVID-19 epidemic resulted in substantial instability in the

economy. During both crises, the analysis found a substantial link with the

price of oil and economic activity, illustrating the severe impact that an

increase in oil prices may have on economic activity.

In 2019, Mo, Chen, Nie, and Jiang conducted a comprehensive

investigation into the influence of crude oil prices on the economic growth of

the BRICS nations, which include Brazil, Russia, India, China, and South

Africa. Their study employed a wavelet-based quantile-on-quantile method,

enabling them to dissect the data into different investor horizons and assess the

overall effects across various quantiles. The research outcomes revealed that

the impact of crude oil prices on economic growth varied significantly among

these countries, fluctuating across different time periods and quantile levels.

37

Digitized by Sam Jonah Library

University of Cape Coast [Link]

These variations stemmed from disparities in oil policies and economic

development strategies unique to each nation. Specifically, in the BRICS

countries, the study identified a positive effect of crude oil prices on economic

growth. However, this positive effect weakened during periods of high oil

prices in Brazil and Russia, and it diminished over time in India. In China, the

research unveiled a positive impact in the short and medium term, followed by

a subsequent negative influence. Interestingly, in the long run, higher crude oil

prices were found to stimulate economic growth in China. Furthermore, the

study highlighted distinct dynamics in South Africa, where a negative effect of

crude oil prices on economic growth was observed in the short term.

Nevertheless, over time, a positive effect re-emerged, although it gradually

waned (Mo et al., 2019).

Zhang, Mou, and Ye (2022) conducted research on the effect of

international crude oil price fluctuations on China's industrial sector, focusing

specifically on dynamic jumps. To achieve this, they used the asymmetric

Autoregressive Moving Average- Exponential Generalized Autoregressive

Conditional Heteroskedasticity (ARMA-EGARCH) model to examine the

features of global price swings and integrated the jump element into the

Autoregressive Moving Average- Exponential Generalized Autoregressive

Conditional Heteroskedasticity-X (ARMA-EGARCH-X) model to explore the

effects of sudden fluctuations on the country’s industrial sector over time. The

researchers grouped abrupt increases in oil prices based on different oil market

conditions and assessed the asymmetrical impact of the shocks. Their results

indicated that global oil price fluctuations display volatility concentration that

is characterized by asymmetry and dynamic jumps. The study found that oil

38

Digitized by Sam Jonah Library

University of Cape Coast [Link]

price jumps had an adverse effect on returns in China's industrial sector, but a

favorable impact on its volatility at the overall level.

Empirical review on Ghana

Dadzie, Nambie, and Obobi (2023) studied the link between the

volatility of Ghana's petroleum and commodities prices. The main objective

was to establish an empirical framework to understand the direction of this

impact. Drawing on the existing literature on time series analysis, the research

explored the ADF, Granger causality, co-integration, vector autoregressive

(VAR), and vector error correction models to assess the connection between

the volatility in petroleum energy prices and the chosen commodity variables.

Findings revealed a significant and enduring connection between petroleum

price volatility and commodity prices in Ghana over 2011 to 2022period.

According to a single equation error correction model, shocks in petroleum

energy prices led to increases in the prices of grains, meat, and cooking oil in

both the short and long term. Additionally, the analysis of impulse response

functions and variance decomposition indicated the presence of both short-

term and long-term associations between these variables.

Nchor, Klepá, and Adamec (2016) investigated the evolving

connection between oil price shocks and crucial macroeconomic indicators in

the economy of Ghana. For this, they used Vector Autoregressive (VAR) and

Vector Error Correction (VECM) models. The inflation rate, and real GDP

growth rate in the and the real effective exchange rate were all factors

addressed. The study emphasized the asymmetric effects of oil price shocks,

suggesting that favorable and unfavorable oil price fluctuations had an

influence on the chosen macroeconomic indicators. The study's empirical

39

Digitized by Sam Jonah Library

University of Cape Coast [Link]

findings demonstrated that both linear and non-linear variations in oil prices

have negative effects on Ghana's macroeconomic indicators. Positive oil price

shocks, in particular, were found to have a greater impact than negative shocks

to government spending, inflation, and the real effective exchange rate.

Industry value added and imports, on the other hand, responded more strongly

to negative oil price shocks. ―Positive oil price shocks accounted for

approximately 30 percent of the fluctuations in government spending, 5

percent in imports, 6 percent in industrial value added, 17 percent in inflation,

and 2 percent in the real effective exchange rate in the long run. Conversely,

negative oil price shocks resulted in roughly 8 percent of changes in

government expenditure, 20 percent in imports, 8 percent in inflation, and 2

percent in the real effective exchange rate over time‖ (Nchor et al., 2016).

Dramani and Frimpong (2020) investigated "the effects of underlying

changes in crude oil prices on the stability of Ghana's macroeconomic

indicators" (Dramani and Frimpong, 2020). They developed a structural vector

autoregressive model to decipher the causes of crude oil market variations.

Their goal was to determine how these shocks affected macroeconomic

variables, and some specificbilateral exchange rates in Ghana. They also

investigated the extent to which the detected shocks influenced the levels of

food and non-food prices. The study's findings revealed that shocks related to

the supply and demand for oil dynamics had a substantial impact on Ghana's

real GDP. Furthermore, the observed shocks had a significant impact on the

bilateral exchange rate between Ghana and the Euro. Furthermore, the findings

implied that very shock examined in the study had a substantial impact on

both food and non-food inflation. This means that oil market disruptions have

40

Digitized by Sam Jonah Library

University of Cape Coast [Link]

a negative impact on inflation, hurting both food and non-food costs in Ghana

(Dramani and Frimpong, 2020).

Oteng-Abayie, Dramani, Sulemana, and Adusah-Poku (2023) did

research titled "The Asymmetric Impact of Oil Price Shocks on Demand for

Goods and Services in Ghana." Their study sought to observe the differential

impacts of favourable and unfavourable changes in oil prices on overall

demand and its diverse constituents spanning from 1970 to 2015. For their

analysis, they used a nonlinear ARDL framework. According to the study's

results, shocks in oil prices within the Ghanaian context had a persistent and

asymmetric influence on overall demand and its constituent parts. More

precisely, a positive oil price shift had a larger positive impact on overall

demand than an unfavourable one from a fall in oil prices. When the various

components of aggregate demand were examined, this trend was similar, with

investment expenditures (0.662) having the most noticeable influence. Based

on the study's suggestions, the research implications imply that policymakers

should consider diversifying their approach to energy consumption. Rather

than just exporting crude oil, government should promote processing and

domestic use of the commodity. Furthermore, the study recommends that

policymakers use hedging tactics and price-smoothing approaches to reduce

the volatility of oil prices (Oteng-Abayie et al., 2023).

In a similar manner, Appiah, Oduro, and Benn (2020) conducted

research on the empirical examination of the influence of oil consumption and

price on Ghana's growth. The study gathered annual time series data,

commencing in 1980 and extending up to 2016, and the researchers used the

Augmented Dickey-Fuller (ADF) test, the Johansen Cointegration test, and the

41

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Ordinary Least Squares (OLS) estimate process. Real GDP per capita, the

amount of crude oil used, the price of crude oil, and the amount spent by the

government are the variables that are utilized. Based on the study's results,

there exists a correlation that is both positive and a noteworthy statistical

relationship was identified between the oil price and long-term

growthnevertheless, there is a correlation that is a substantial relationship

between crude oil usage and long-term GDP growth. (Appiah et al., 2020).

Chapter Summary

The chapter reviewed literature on crude oil price volatility and

industrial output. The theoretical review presented hypotheses that explain

how the volatile nature of oil prices might have a ripple effect on the

economy. In addition, the empirical review demonstrated the models that were

utilized by other researchers, as well as the empirical findings about the effect

that fluctuations in the price of crude oil have on countries and the gap that

exists as summarised in Table 1. The study therefore, intends to consider the

industrial sector of Ghana, the second highest contributor to the country’s

GDP and the second highest consumer of refined crude oil products to

determine how crude oil price volatility may affect its output.

The current study addresses theoretical, methodological and empirical

gaps in line with the nexus between crude oil price volatility and industrial

sector output. Theoretically, the nexus between crude oil price volatility and

the industrial sector output can be heterogeneous and adaptive (Asafo-Adjei,

Adam & Darkwa, 2021), in line with the heterogeneous market hypothesis

(Müller et al., 1997) and the adaptive market hypothesis (Lo, 2005). However,

prior studies conducted on the impact of other macroeconomic factors on the

42

Digitized by Sam Jonah Library

University of Cape Coast [Link]

performance of the industrial sector failed to address the extent to which the

nexus is heterogeneous and adaptive (see, Humaira et al., 2019; Iganiga et al.,

2021; Jiranyakul, 2016; Tunyo et al., 2021) thereby revealing their asymmetric

outcomes. This is because these studies employ methodologies that do not

account for the heterogeneous and adaptive behaviours of the industrial sector

output.

Overall, existing studies that have investigated the impact or effects of

OPV have looked at it concerning macroeconomic indicators. The academic

literature on economics is replete with research that investigate the connection

between OPV and economic growth without limiting their focus to a particular

sector of the economy. To the best of my knowledge, the topic on the impact

of OPV on Ghana's industrial sector is almost entirely absent from the relevant

body of academic research. As a result, this creates a gap, which in turn makes

it necessary to do this research.

Hence, the current study utilises the quantile regression approaches as

adopted by prior studies (Adebayo, Rjoub et al., 2022; Archer et al., 2022;

Barson et al., 2022; Demir et al., 2020) to divulge the conditions of the market

in terms of stress, normal and boom. It also employs the bi-wavelet approach

which explores co-movements and lead-lag nexus between time series data

across time and frequency (Armah, Amewu & Bossman, 2022; Asafo-Adjei,

Adam & Darkwa, 2021; Singh, Bansal & Bhardwaj, 2022). Therefore, the

purpose of this study is to bridge the knowledge gap by investigating the

effects of OPV on industrial sector output in particular, measuring the

asymmetric relationship between crude oil price volatility and industrial sector

output, determining the time and frequency connectedness between crude oil

43

Digitized by Sam Jonah Library

University of Cape Coast [Link]

price volatility and industrial sector output, and the conditional causality from

crude oil price volatility to industrial sector output. Table 1 presents summary

and gaps in literature.

44

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Table 1: Summary and Relevance

48

Digitized by Sam Jonah Library

University of Cape Coast [Link]

49

Digitized by Sam Jonah Library

University of Cape Coast [Link]

50

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

CHAPTER THREE

RESEARCH METHODS

Introduction

This section provides a comprehensive account of the research

methodology, including the research paradigm, approach, and the empirical model

specification. Also, this section describes the variables, the associated variable

measurements, data sources, and the model estimation methodology.

Research Philosophical Perspectives or Paradigm

Holden and Lynch (2004) state that philosophical perspectives serve as the

underlying structures that encompass all academic research. The research

paradigm is a comprehensive framework that incorporates fundamental theories,

crucial topics, and high-quality research methodologies for obtaining solutions

(Hennink, Hutter, and Bailey 2020; Cameron 2009). The two primary

philosophical stances in social science are known as ontology and epistemology

(Ormston, Spencer, Barnard, & Snape, 2014). Epistemology is the study of how

knowledge is recognized within a certain field. In addition to these moderate

approaches, positivism and interpretivism represent two poles in the field of

epistemology. The positivist school of thought in science accepts as true only

objective claims.

The positivist model ensures the researcher and the respondent are kept at

arm's length from one another in the pursuit of value-free results (Ormston,

Spencer, Barnard, & Snape, 2014.). Not only that, but literature often links

positivist paradigms with quantitative research methods (Smith, 1983).

51

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Quantitative research has a large focus on numbers and enables the creation and

testing of hypotheses. The Positivist paradigm also follows the deductive

approach, which looks at a particular theory, formulates a hypothesis from that

theory, and then tests the hypothesis to approve or disprove the theory (Slevitch,

2011). Because of its emphasis on numerical data and its desire to test hypotheses

based on theoretical models, this study adheres to the positivist paradigm of

epistemology.

Research Design

A study's research strategy may be affected by the research design

chosen for the investigation as a whole. The research design is the framework

for the study and the starting point from which all other decisions regarding the

study, its execution, and its results will be made (Kothari, 2004). Saunders et al.

(2012) identify three primary types of research studies: descriptive studies,

exploratory studies, and explanatory studies. This study took on an explanatory

research strategy since its goal was to assess the influence of one or more

independent factors on a second, more important variable. A crucial part of the

process of creating and validating theoretical models is using causal research

design. This study is based on a causal research design since its authors set out

to ascertain whether or not the effect of crude oil price volatility differs across

different industries and whether or not there is a temporal and/or spatial

correlation between the two.

52

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Research Approach

Research can be characterized as quantitative, qualitative, or a

combination of the two. A qualitative work applies to studies that concentrate on

naturally occurring phenomena and in natural settings. It draws conclusions about

mathematically inappreciable occurrences based on intuitive and perceptual

observations. Blending quantitative and qualitative methods into a single strategy

is what the mixed approach is all about. It explains the research challenge from

both the quantitative and qualitative vantage points in depth (Creswell, 2013). In

order to draw conclusions about the course of action of a phenomenon, the mixed

research technique combines elements of qualitative and quantitative

methodologies (Creswell & Clark 2011). To do this, the current study takes a

quantitative approach and attempts to measure the correlation between its

independent variables.

The focus of the quantitative method is on quantitative measurement and

the use of numerical analysis of data for elucidation purposes. As the collected

data can be simply analyzed with common statistical tools, a quantitative

methodology will be used for the study. Meanwhile, the quantitative strategy

makes use of numerical and measurable data in its strategies, measurements, and

designs (Simon, Lee, Cottrell & Verleysen, 2007). In addition to proper

measurement of the variables under study, the design relies on the concepts of

demonstration, substantiation, and confirmation verifiability. The study is purely

quantitative. This is because it deploys quantitative variables both dependent and

independent to study the effect of crude oil volatility on the output of the

53

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

industrial sector and time and frequency connectedness of the industrial sector

and crude oil prices volatility.

Data Sources

The study analyzed the effect of crude oil price volatility on the output of

Ghana's industrial sector using secondary data for the period of time spanning

from 2001 to 2020. The data were in the form of a monthly time series. The price

of crude oil, the output of industry, mining and quarrying, manufacturing, oil and

gas, construction, water and sewage, and electricity are some of the elements that

are taken into account. The research also adjusted for other variables, such as the

amount of energy consumed, foreign direct investment, and interest rate. The

Energy Commission, the Bank of Ghana (BOG), the International Monetary Fund

(IMF), the Ghana Statistical Service (GSS), and the World Bank's database of

World Development Indicators (WDI) were the sources of the data.

Theoretical Model Specification

The research creates a theoretical model of the relationship between crude

oil price volatility and industrial sector production. using ―the symmetric/linear

relationship theory of growth‖ (Hamilton, 1983; Hooker, 1986). This theory

argues that fluctuations in oil prices causes changes in economic growth (Taofik,

2018). That is, a surge in the price of oil would therefore lead to a drop in GDP

(Taofik, 2018). Based on this theory, the current study presents a theoretical

framework of how the price of oil affects economic growth as follows;

(1)

where is economic growth, and is oil price.

54

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

The study follows prior literature to analyse the effect of volatility in oil

price on growth (Aimer, 2016; Akalpler et al., 2018; Al-Sasi et al., 2017). Oil

price volatility is a measure of how much oil prices change over time, whereas oil

price is the actual level of prices at a particular point in time (Taofik, 2018).

Higher oil price volatility means that oil prices are more unstable and subject to

larger fluctuations, while lower oil price volatility indicates greater stability and

predictability in oil prices (Chang, Baloch, Saydaliev, Hyder & Dilanchiev,

2022). Hence, equation 1 can be modified as;

(2)

where is output, and ROVX is realised crude oil price volatility.

Empirical Model Specification

Estimation Techniques

Based on empirical literature, the effect of crude oil price volatility on the

output of the industrial sector was specified using econometric techniques taking

into consideration market and economic situations, and time and frequency

connectedness. This is relevant to divulge the heterogeneous and adaptive

behavior of the market and performance. In this regard, the quantile regression,

the conditional causality-in-quantiles test and wavelet approaches are employed.

Quantile regression model

The study used the quantile regression approach to examine the

relationship between time series data. It shows the effect of one variable

(independent variable) on the conditional distribution of another variable

(dependent variable). This technique is applicable when the time series data is not

55

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

normally distributed, and hence, the coefficient from the mean equation of the

Ordinary Least Square (OLS) technique would not be reliable (Koenker &

Bassett, 1978). Hence, quantile regression is adopted to assess the asymmetric

effect of one variable on the quantiles of another variable. The quantile regression

model is shown as

(3)

where denotes output of the industrial sector at time t, represents realised

crude oil price volatility at period t. Moreover, θ is the θth quantile of the

regressors and β represents parameters to be estimated at each quantile.

Existing studies such as Adebayo, Rjoub, Akinsola and Oladipupo (2022),

Archer, Owusu Junior Adam, Asafo-Adjei and Baffoe (2022), Barson, Owusu

Junior, Adam and Asafo-Adjei (2022), Boateng, Adam and Owusu Junior (2021),

Demir, Pesqué-Cela, Altunbas and Murinde (2020), among others made use of the

quantile regression approach. The quantile regression technique, introduced by

Koenker and Bassett in the 1970s, models the conditional quantile of a response

variable as a linear function of the explanatory variables, in contrast to the

traditional practice of using only the conditional mean. This approach yields more

reliable estimates in the presence of outliers in the response variable. Moreover,

quantile regression provides a detailed understanding of the impact of the

independent variable on the response variable.

Examining correlations between realised crude oil volatility and returns on

output of the industrial sector occurred across 19 different quantiles, ranging from

the 0.05th to the 0.95th quantile. These quantiles were chosen in order to determine

56

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

whether changes in the price of realised crude oil would also affect the industrial

sector's output. Hence, based on the quantiles of the study, three varying

economic conditions are used. They are stress (0.05-0.35), normal (0.40-0.65) and

boom (0.70-0.95) as advocated by prior studies (Adebayo et al., 2022; Archer et

al., 2022; Demir et al., 2020).

As a result, estimates using quantile regression are more robust to outliers

in the response measurement. In addition to this, quantile regression provides a

more in-depth look at a regressand's dependence on an independent variable. In

other words, it comprehensively describes and characterizes the data by

displaying the effects of the regressor on the explanatory variable over the range

of the dependent variable. To illustrate how the two variables are related, this is

done. In most cases, the equation that best describes the quantile regression model

is as follows:

(4)

where represents the vector of unknown parameters related with the θth

quantile. The quantile regression minimizes | + | , as a result,

the sum provides the asymmetric penalties | for underprediction and

| for overprediction. The optimization problem outlined below can be used to

determine the coefficient or the quantile estimator.

min ∑ | - |+∑ | - | (5)

where is the dependent variable and is a K by 1 vector of regressors.

Examining correlations between realised crude oil volatility and returns on

output of the industrial sector occurred across 19 different quantiles, ranging from
57

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

the 0.05th to the 0.95th quantile. These quantiles were chosen in order to determine

whether changes in the price of realised crude oil would also affect the industrial

sector's output. Hence, based on the quantiles of the study, three varying

economic conditions are used. They are stress (0.05-0.35), normal (0.40-0.65) and

boom (0.70-0.95) as advocated by prior studies (Adebayo et al., 2022; Archer et

al., 2022; Demir et al., 2020).

Empirical objective one

This objective estimates the asymmetric relationship between realised

crude oil price volatility and industrial sector output. After controlling for interest

rate, energy consumption, crude oil price and foreign direct investment, the

equation can be specified as;

(6)

where denotes industrial sector output at time t, represents

realised crude oil price volatility at period t. Again, , , and are the

control variables representing crude oil price, interest rate, energy consumed and

foreign direct investment respectively. Moreover, θ is the θth quantile of the

regressors and β represents parameters to be estimated at each quantile.

Nonlinear causality-in-quantiles approach

The study utilizes the nonlinear causality-in-quantile approach as

suggested by Balcilar, Gupta and Pierdzioch (2016) to explore the causality

between two variables. Hence, to assess causality at diverse distributions of a

58

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

variable for time series data that are non-normally distributed and non-linear, the

causality-in-quantile test is preferred (Archer et al., 2022; Hammoudeh &

Roubaud, 2019; Jena, Tiwari, Jeong, Härdle & Song, 2012). The study tests that

is not caused by at various -quantile based on the lag-vector of

{ } if

| | (7)

Nonetheless, causes in the -quantile based on

{ } if

| | (8)

where represents the -quantile of . It must be noted that the

conditional quantiles of , rely on whereas the quantiles range

between and .

By defining the vectors , , and

. Hence, functions and are the

conditional distribution functions of conditioned on vectors and

correspondingly. Accordingly, as indicated by Jena et al. (2019), the causality-in-

quantile hypothesis regarding equations (9) and (10) can be shown as:

{ } (9)

{ } (10)

Equation (9) (H0 – Null Hypothesis):

Where:

P {} represents the probability.

F(yt|γ(t-1)) is a conditional probability distribution of variable yt given γ(t-1).

59

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Qθ(Y(t-1) |γ(t-1)) is another conditional probability distribution of variable Y(t-1)

given γ(t-1).

is a parameter.

Equation 9 is essentially stating the null hypothesis (H0). It’s saying that the

probability of the event F(yt|γ(t-1)) Qθ(Y(t-1) |γ(t-1)) being equal to θ is equal to

1. Suggesting that there is no significant relationship or difference between the

two conditional probability distributions F(yt|γ(t-1)) and Qθ(Y(t-1) |γ(t-1)).

Equation (10) (H1 - Alternative Hypothesis):

Where:

P {} represents the probability.

F(yt|γ(t-1)) is, once again, a conditional probability distribution of variable yt

given γ(t-1).

Qθ(Y(t-1) |γ(t-1)) is the conditional probability distribution of variable Y(t-1)

given γ(t-1).

θ is a parameter.

Equation 10 is the alternative hypothesis (H1). It suggests that the probability of

the event F(yt|γ(t-1)) Qθ(Y(t-1) |γ(t-1)) being equal to θ is less than 1. This

implies that there is evidence to suggest that there might be a significant

relationship or difference between the two conditional probability distributions

F(yt|γ(t-1)) and Qθ(Y(t-1) |γ(t-1)).

Empirical objective two

This objective determines the conditional causality of realised crude oil price

volatility to industrial sector output.

60

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

The study tests that is not caused by at various -quantile

based on the lag-vector of { } if

| (11)

Nonetheless, causes in the -quantile based on

{ } if

| (12)

where denotes industrial sector output at time , represents

realised crude oil price volatility at period .

Bi-wavelet model

The bi-wavelet approach is presented to find out the time and frequency

connectedness between two time series data. This approach is necessary in

deciphering the calendar time and horizon (short, medium and long-terms)

resource management decisions are useful depending on the kind of directional

connectedness between two time series data. Since the bi-wavelet is non-

parametric bivariate, it is impossible to control for any other variable. Also, this

method can be used irrespective of the assumption(s) about the time series data.

Torrence and Compo (1998) define the wavelet transform coherence

(WTC) as the normalization of the squared cross-absolute spectrum. The squared

wavelet coherence is denoted as

| |
(13)
| |

61

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

where and shows co-movements between two variables, is a smoothing

factor and the square difference ranges from 0 and 1, wherein a value near to 0

indicates a poor co-movement, whereas a number close to 1 indicates a significant

co-movement. The WTC phase difference depicts the oscillation's disruptions.

Bloomfield, McAteer, Lites, Judge, Mathioudakis, and Keenan (2004) analyze the

phase difference between x(t) and y(t).

{ }
( { }
) (14)

where ℑ and ℜ stand for imaginary and real operators, respectively.

In graphical terms, the arrows pointing to the right (left) represent in-phase

(out-of-phase) time series. A downward-pointing arrow suggests that the second

time series trails the first by ⁄ , whereas an arrow pointing upward indicates

that the first time series trails the second by ⁄ . Places with a lot of co-

movements are shown by a red (warm) while areas with fewer co-movements are

indicated by a blue (cool). An extended methodology of the bi-wavelet approach

can be found in the works of Armah, Amewu and Bossman (2022), Asafo-Adjei,

Adam and Darkwa (2021), Singh, Bansal and Bhardwaj (2022), among others.

Empirical objective three

This objective quantifies the responsiveness of the industrial sector output to

realised crude oil price volatility across time and frequency.

| |
(15)
| |

where and shows industrial sector output and realised crude oil

price volatility.

62

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Variable Descriptions and Sources of Data

Table 2 presents codes, names of variables, unit of measurement, data

sources and a priori. The variables listed in the table below were used to analyse

the nexus between crude oil price volatility and industrial sector output in Ghana.

63

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Table 2: Description of Variables, Units, and Sources of Data

64

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Data Processing and Analysis

The Chow and Lin (1971) extrapolation technique was used in conjunction

with the E-Views statistical package to produce the monthly time series for the

industrial sector, the manufacturing sub-sector, the mining and quarrying sub-

sector, the construction sub-sector, the water and sewage sub-sector, the

electricity sub-sector which were available in quarterly form, and the foreign

direct investment which was in an annual form. Because there are no restrictions

on the sub-periods that can be used when disaggregating data with this approach,

there is no danger to the reliability of the findings when employing it. Using

monthly time series data for this study is justified as it aligns with the frequency

of available data on crude oil prices and ensures that we can capture more detailed

variations in both crude oil prices and industrial output over time, allowing for a

more precise examination of the connection with crude oil price volatility and

industrial output in Ghana. Additionally, the conversion of quarterly industrial

output data to a monthly frequency allows for consistent temporal alignment and

reduces potential data inconsistencies that might arise from using mismatched

time intervals.

The data was further processed using the natural logarithm of the time

series data except for the realised crude oil volatility which was extracted in

logarithmic returns through the GARCH (1,1) process. To aid in the descriptive

analysis, the study employed charts such as minimum, maximum, median, mean,

Kurtosis, Skewness, standard deviation (Std. Dev), Jarque-Bera statistic,

Kwiatkowski–Phillips–Schmidt–Shin test (KPSS), Pearson's r and the Teraesvirta

65

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

neural network (TRSNN) were used to find a correlation between the two

variables. In particular, we utilize skewness and kurtosis to look at the

distribution's overall symmetry (asymmetry) and peakedness (flatness),

respectively.

Both the skewness and kurtosis are combined to assess normality of the

data. This is hypothetically confirmed by the Jarque-Bera statistic based on the

5% significance level. The KPSS test was subsequently used to assess issues of

stationarity. The Teraesvirta Neural Network (TRS) test was used to address

linearity of the data. Correlation matrix is shown to ascertain the direction and

magnitude of variables’ connectivity as well as determine their degree of

collinearity. The mean, standard deviation (Std. Dev), Kurtosis, Skewness,,

Jarque-Bera statistic were estimated using the EViews -10 statistical software.

The Kwiatkowski-Phillips-Schmidt-Shin (KPSS) and the Teraesvirta Neural

Network (TRSNN) method were analysed through the R programming statistical

software.

The three research questions were answered using the quantile regression,

conditional causality in quantiles and bi-wavelet approaches respectively. The

quantile regression was used assess the effect of realised crude oil price volatility

on varying economic conditions of the industrial sector (Adebayo et al., 2022).

The economic conditions considered in this study as also supported by prior

studies are stress, normal and boom economic conditions (Adebayo et al., 2022;

Archer et al., 2022; Barson et al., 2022; Boateng et al., 2021; Demir et al., 2020).

The causality in quantiles test was used to investigate the conditional causality

66

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

from crude oil price volatility to industrial sector output. The bi-wavelet was also

employed to examine the specific time and frequency (investment horizon of

short, medium and long-terms) at which either crude oil or industrial sector

variable leads/or lag the other. The main estimations were performed using the R

programming statistical software.

Chapter Summary

This chapter detailed the research methods that were followed to complete

the study. This study took a quantitative methodology, grounded on the positivist

research paradigm. The study employed an explanatory research methodology

because its researchers wanted to know how oil price volatility affected Ghana's

industrial sector. This research made use of secondary data collected on a monthly

basis from 2001 to 2020. The study quantitatively examined the influence of

realised crude oil volatility on the industrial sector of Ghana. This effect was

executed across economic conditions (stress, normal and boom) as well as time

and frequency. For this reason, the quantile regression and bi-wavelet approaches

were utilised as the study’s estimation techniques. A preliminary data analysed

was investigated to first assess the time series data in terms of normality,

stationarity, and linearity justifying the use of robust techniques that could explain

the dynamics of the nexus.

67

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

CHAPTER FOUR

RESULTS AND DISCUSSION

Introduction

This research seeks to analyse the nexus between realised crude oil price

volatility and industrial sector output at consolidated and segmented levels in

Ghana. It employed five sub-divisions within Ghana's industrial sector,

comprising mining and quarrying, manufacturing, electricity generation, water

treatment, and construction. The industry sector aggregate was added in this study

for further inferences. The findings of this study were presented in this chapter

and discussion following a quantitative research methodology coupled with an

explanatory research framework. Accordingly, the study employs a secondary

source data between 2001 to 2020 which was a monthly time-series data. The

chapter's organization is structured in the following manner: First, descriptive

statistics of the monthly time series data is performed. Second, the main

estimations are performed using the quantile regression, bi-wavelet and

conditional causality in quantiles approaches.

Descriptive Statistics

The descriptive statistics is made up of realised crude oil price volatility

(ROVX), crude oil price (OP), five industrial sector divisions and the industrial

sector output. These were construction (CONS), electricity (ELECT),

manufacturing (MFG), mining and quarrying (MQ), and water and sewerage

(WS). The industry sector output (ISO) was also included. Four control variables

were employed to supplement the quantile regression estimation. These are

68

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

energy consumption (EC), foreign direct investment (FDI), interest rate (INT),

and crude oil prices (OP). The variables used in the descriptive statistics are in

natural logarithm with exception to the crude oil price volatility which is in its

natural logarithmic returns extracted through the GARCH (1,1) model. The

transformation of crude oil price volatility into natural logarithmic returns through

the GARCH (1,1) model is crucial in this empirical analysis for several reasons.

Natural logarithmic returns are preferred in financial and economic studies as they

capture percentage changes in price, making them a more meaningful

representation of asset price dynamics. Additionally, the GARCH (1,1) model

specializes in modeling volatility, and expressing conditional variances as natural

logarithmic returns facilitates the interpretation of volatility changes. Moreover,

using natural logarithmic returns enhances the statistical properties and

interpretability of the variable, ensuring that it aligns better with the assumptions

required for accurate regression analysis in the study of its impact on industrial

output in Ghana.

Minimum, maximum, median, mean, standard deviation (Std. Dev),

skewness (SKS), kurtosis, Jarque-Bera, Kwiatkowski–Phillips–Schmidt–Shin test

(KPSS) and Teraesvirta Neural Network (TRSNN) tests were performed to first

ascertain the nature, distribution, stationarity and linearity of the time series data.

Particularly, the skewness and kurtosis are respectively used to examine the

degree of symmetry (asymmetry) and peakedness (flatness) of the distribution.

Both the skewness and kurtosis are combined to assess normality of the data.

Standard deviation values closer to zero indicates less dispersion within the

69

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

dataset. Skewness values that are closer to zero denotes a symmetrical

distribution, otherwise asymmetrical distribution (positively or negatively

skewed). Also, kurtosis values between -0.5 and +0.5 suggests a mesokurtic

distribution. On the other hand, Kurtosis values above 0.5 means a leptokurtic

distribution, but a platykurtic distribution if less than -0.5.

This is hypothetically confirmed by the Jarque-Bera statistic with a null

hypothesis of normality. The KPSS test is subsequently used to assess issues of

stationarity with a null hypothesis of a stationary series. The Teraesvirta Neural

Network (TRSNN) test is used to address linearity of the data with a null

hypothesis of a linear relationship. All the hypothetical tests are assessed at the

5% significance level. The number of observations (N) of the variables are also

presented in Table 3.

70

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

Table 3: Descriptive Statistics

Min Max Median Mean Std. Dev. SKS Kurtosis Jarque-Bera KPSS TRSNN N
CONS 6.361 10.274 8.464 8.402 1.325 -0.192 1.499 24.023** 4.885** 97.861** 240
EC 6.326 7.299 6.811 6.772 0.227 -0.155 2.732 1.685** 4.236** 9.194** 240
ELECT 5.411 8.552 6.858 6.974 1.057 0.148 1.580 21.032** 4.859** 39.084** 240
FDI 17.892 22.079 21.677 20.953 1.356 -1.131 2.591 52.883** 3.465** 49.170** 240
INT 3.042 3.866 3.311 3.339 0.173 1.243 4.575 86.568** 2.238** 2.418 240
ISO 8.433 11.765 10.171 10.148 1.057 -0.029 1.614 19.249** 4.917** 58.151** 240
MFG 7.614 10.792 9.344 9.271 0.908 -0.081 1.857 13.317** 4.876** 3.311 240
MQ 7.404 10.768 9.040 9.029 1.127 0.100 1.527 22.092** 4.844** 31.942** 240
OP 2.911 4.888 4.128 4.066 0.492 -0.367 2.277 10.621** 1.504** 2.602 240
ROVX 0.004 0.167 0.007 0.011 0.017 6.932 58.950 33225.970** 0.265 11.906** 240
WS 4.809 8.113 6.351 6.368 1.042 0.150 1.625 19.812** 4.853** 11.710** 240
Source: Author’s Construct (2023)

Note: The variables CONS, EC, ELECT, FDI, INT, ISO, MFG, MQ, OP, ROVX and WS represent Construction, Energy consumption,
Electricity, Foreign Direct Investment, Interest rate, Industry Sector Output, Manufacturing, Mining and Quarrying, Crude oil price,
Realised crude oil price volatility, and Water and Sewerage respectively. Also, **, and * denotes significance at 1% and 5%
respectively. Min, Max, Std. Dev., SKS, KPSS, TRSNN, and N denote Minimum, Maximum, Standard deviation, Skewness,
Kwiatkowski–Phillips–Schmidt–Shin test (KPSS), Teraesvirta Neural Network, and number of observations of the data respectively.

71

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Table 3 illustrates the minimum, maximum, median and mean values

of the variables. FDI (min; 17.892, max; 22.079, median; 21.67, mean;

20.953) has the highest values for all four statistics (minimum, maximum,

median and mean), indicating that it has the highest overall values compared

to the other variables. ROVX has the lowest values (min; 0.004, max; 0.167,

median; 0.007, mean; 0.011) for all four statistics, indicating that it has the

lowest overall values compared to the other variables. This is not surprising

because logarithmic returns were used for ROVX relative to the natural

logarithm of the remaining variables. For most variables, the mean and median

are relatively close, suggesting that the distribution of values is roughly

symmetric. However, for INT, the mean is slightly higher than the median,

suggesting that there may be some outliers on the high end of the distribution.

The ranges between the minimum and maximum values vary widely across

variables, from a narrow range of 0.163 (ROVX) to a wide range of 4.187

(FDI).

Furthermore, from Table 3, the standard deviation values closer to zero

means that the data points of the variables are less disperse. For instance, a

standard deviation value of 0.227 for EC, 0.173 for INT, 0.492 for OP and

0.017 for ROVX. The standard deviation values for these variables

approaching zero illustrate that these variables are less dispersed. It is also

observable that the variables are closer to symmetry as the skewness values

approach zero except for ROVX, FDI and INT that depart from symmetry.

The Kurtosis values for all variables above 0.5 indicates that the variables

have a leptokurtic distribution. Since there is no simultaneous symmetry and

mesokurtic distribution for each variable, they are considered not normally

72

Digitized by Sam Jonah Library

University of Cape Coast [Link]

distributed. This is confirmed by the Jarque-Bera statistic with a null

hypothesis of normality. Because the Jarque-Bera statistic for all the time

series data are significant at the 1% level, the null hypothesis is rejected to

mean that the series are not normally distributed.

Additionally, it can be seen from the KPSS test with a null hypothesis

of a stationary series at a 1% level, the null hypothesis of the test is rejected

except for ROVX. Hence, almost all the series are not stationary. Moreover, to

investigate linearity of the variables, the Teraesvirta Neural Network (TRS)

test with a null hypothesis of linearity is used. It can be confirmed that at the

1% level the null hypothesis of the series (except for INT, MFG and OP) is

rejected. This indicates that most of the series are not linear. As shown, the

non-normal distribution, non-stationary series and the non-linear series permit

the application of the quantile regression and bi-wavelet approaches since they

are robust in such circumstances as confirmed by prior studies (Adebayo et al.,

2022; Archer et al., 2022; Armah, Amewu & Bossman, 2022; Demir et al.,

2020; Singh, Bansal & Bhardwaj, 2022).

The time series presentation of the main variables for this study are

presented in Figure 7. Since, ROVX obtained through the GARCH (1,1)

model discussed in appendix A is in logarithmic returns, and the industrial

sector output variables are in natural logarithm, the series are presented

separately, and their patterns are observed across time.

73

Digitized by Sam Jonah Library

University of Cape Coast [Link]

14 0.18

Natural Logarithm of Industrial Sector Output

0.16
12
0.14
10

Returns for ROVX

0.12

8 0.1

6 0.08

0.06
4
0.04
2
0.02

0 0
Jan-01
Jan-02
Jan-03
Jan-04
Jan-05
Jan-06
Jan-07
Jan-08
Jan-09
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Jan-18
Jan-19
Jan-20
MQ MFG ELECT WS CONS ISO ROVX

Figure 7: Time series plots of ROVX and Industrial Sector Output

Source: Author’s Construct (2023)

Note: The variables, MQ, MFG, ELECT, WS, CONS, ISO, and ROVX
represent Mining and Quarrying, Manufacturing, Electricity, Water and
Sewerage, Construction, Industrial Sector Output, and Realised crude oil
price volatility, respectively.

Figure 7 depicts observable shocks in ROVX over time, with major

fluctuations during periods of economic crises, such as the 2008 Global

Financial crisis, the British exit from the European Union (BREXIT) in 2016,

and the COVID-19 Pandemic in 2020. In contrast, the natural logarithm of the

industrial sector output trends upward, despite experiencing occasional

downturns during these same crisis periods. Thus, an increase in crude oil

volatility shocks can be observed with a decline in certain sub-sectors such as

construction, manufacturing, and mining and quarrying. On the other hand, a

rise in electricity and water and sewerage sector output can be observed during

BREXIT. Additionally, Figure 7 illustrates a slight decline in the industrial

74

Digitized by Sam Jonah Library

University of Cape Coast [Link]

sector output amidst the COVID-19 Pandemic. Therefore, to understand the

potential opposing movements between crude oil volatility and some industrial

sector outputs, inferential analysis must be performed to assess the

susceptibility of the industrial sector to shocks from ROVX.

Correlation Matrix

The correlation matrix (see Appendix B) shows the level of

relationship between the variables. The variables used in the correlation matrix

are in natural logarithm with exception to the crude oil price volatility which is

in its natural logarithmic returns. Direction and size of the relationship are

both made clear. To assess potential issues related to multicollinearity, it is

essential to conduct a correlation test. This test serves as a means to identify

and address concerns arising from multicollinearity in the study's variables. It

is recommended that the correlation coefficient between two independent

(explanatory) values be less than 0.8, as suggested by prior research (Gujarati

& Porter, 2009; Obite, Olewuezi, Ugwuanyim, & Bartholomew, 2020).

Interest rate is negatively correlated with all the variables (see

Appendix B). The connection between crude oil price and the industrial sector

is positive and significant. This implies that crude oil price is a possible

predictor of the industrial sector and can be treated as part of the selected

control variables. The realised crude oil price volatility (ROVX) is less

connected to the industrial sector as shown from the magnitude of the

correlation coefficients depicting weak correlations. This is not startling

because the realised crude oil volatility forms part of the crude oil price as

shocks which seldomly occur and might not entirely reflect the behavior of

other markets across all economic conditions as well as time and frequency.

75

Digitized by Sam Jonah Library

University of Cape Coast [Link]

It must be noted that the high correlation values (above 0.8) between

the dependent variables (industrial sector and sub-sectors) as well as between

the dependent and explanatory variables (in the cases of CONS and EC;

CONS and FDI; MFG and FDI) do not raise multicollinearity issues. Hence,

multicollinearity becomes a problem when the explanatory variables are rather

highly connected as advocated by prior studies (Nkrumah-Boadu, Owusu

Junior, Adam, & Asafo-Adjei, 2022; Obite, Olewuezi, Ugwuanyim, &

Bartholomew, 2020). Accordingly, the relationships between the explanatory

variables with correlation coefficients less than 0.8 can be considered ideal

with no issues of multicollinearity in the quantile regression model of this

study.

The Asymmetric Relationship between Realised Crude Oil Price

Volatility and Industrial Sector Output
Asymmetric Model Results

This section shows estimates from the quantile regression approach on

the effect of realised crude oil price volatility (ROVX) on the industrial sector

output. The industrial sector output includes the five divisions of the industrial

sector (construction (CONS), electricity (ELECT), manufacturing (MFG),

mining and quarrying (MQ), and water and sewerage (WS)) and the

aggregated industrial sector output (ISO). Four control variables were

employed to supplement the quantile regression estimation. These are energy

consumption (EC), foreign direct investment (FDI), interest rate (INT) and

crude oil price (OP).

Estimates are presented for 19 quantile (τ) distributions of industrial

sector output from 0.05 through to 0.95. The lower quantile range of 0.05-0.35

(which represents stressful economic conditions), the middle quantile of 0.40-

76

Digitized by Sam Jonah Library

University of Cape Coast [Link]

0.65 (which represents normal economic conditions), and the higher quantile

of 0.70-0.95 (which represents economic boom) are used to illustrate

economic conditions of industrial sector output (see, Adebayo et al., 2022;

Archer et al., 2022; Demir et al., 2020). The effect of realised crude oil price

volatility (ROVX) on the economic conditions of the construction industry

(CONS) is presented in Table 4.

Table 4:Quantile Regression Estimates (ROVX on CONS)

Construction Sub-sector
Τ OP ROVX EC INT FDI
Stress
0.05 -0.2142 0.3292 -0.8838 0.4430 0.6035***
0.10 -0.7599 -2.1002 -0.3254 -0.2283 0.6424***
0.15 -1.0577** -4.8242 0.6048 -1.3345 0.5880***
0.20 -1.5771*** -6.5571 1.4963*** -2.7285*** 0.6375***
0.25 -1.3201*** -6.0308 1.3928*** -2.7390*** 0.6348***
0.30 -1.3184*** -6.3169 1.3426*** -2.7590*** 0.6567***
0.35 -1.1205*** -6.0773 1.3764*** -2.8708*** 0.6326***
Normal
0.40 -1.0429*** -6.0172 1.3675*** -2.9213*** 0.6319***
0.45 -1.0299*** -6.3005 1.3530*** -2.9455*** 0.6408***
0.50 -0.9837*** -6.7112 1.4501*** -3.0548*** 0.6208***
0.55 -0.9814*** -4.8618 1.4794*** -3.1215*** 0.6231***
0.60 -0.9711*** -4.6634 1.3860*** -3.0530*** 0.6422***
0.65 -0.8919*** -3.8564 1.2649*** -2.9357*** 0.6494***
Boom
0.70 -0.7955*** -4.0716* 1.4056*** -2.8595*** 0.5770***
0.75 -0.7243*** -3.7101* 1.3453*** -2.6657*** 0.5551***
0.80 -0.6658*** -3.3303** 1.2376*** -2.5742*** 0.5660***
0.85 -0.6552*** -3.3506** 1.2067*** -2.5593*** 0.5728***
0.90 -0.6036*** -3.0785* 1.1043*** -2.4660*** 0.5836***
0.95 -0.4011*** -0.5511 1.0427*** -2.3587*** 0.5498***
Source: Author’s Construct (2023)

Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

As presented in Table 4, it is observable that the effect of ROVX on

the output of the construction sub-sector is negative. Therefore, ROVX

transmits negative shocks to the construction industry. The negative effect of

77

Digitized by Sam Jonah Library

University of Cape Coast [Link]

ROVX on boom economic conditions (quantiles 0.7-0.90) of the construction

sub-sector output is significant. This implies that the realised crude oil

volatility has a substantial effect on the construction sub-sector when the

economic conditions are favorable. In other words, extreme good performance

of the construction sub-sector attracts negative shocks from the realised crude

oil. On the other hand, at stressed (quantiles 0.05-0.35) and normal economic

conditions (quantiles 0.40-0.65) of the construction sub-sector, there is

insignificant nexus between the construction sector and realised crude oil price

volatility.

Moreover, from Table 4, the control variables have significant effects

on the construction sector at most quantiles of the construction sector. For

instance, rising crude oil prices have a depressing effect on construction sub-

sector in down, steady, and up markets from quantiles 0.15 to 0.95. However,

investors of the energy market and the construction sector may find the

markets attractive with the quest of offering diversification benefits. On the

other hand, energy consumption has a positive effect on the construction sub-

sector from quantiles 0.20 to 0.95. Hence, inadequate energy supply is

expected to push up crude oil price which will increase the production cost of

the construction sub-sector, and thereby reducing output. Inadequate energy

supply and inefficient energy consumption, as argued by Kassim and Isik

(2020), mitigate industrial expansion, warranting cautious observation of the

negative nexus at all quantiles.

The nexus between interest rate and the construction sub-sector is

negative and significant for quantiles 0.20 to 0.95. Favorable growth in the

construction sub-sector is correlated with increased energy consumption and

78

Digitized by Sam Jonah Library

University of Cape Coast [Link]

foreign direct investment (FDI). Beji and Belhadj (2014) highlighted the long-

term benefits of industrialization, including increased economic diversity, the

spread of new technologies, lower unemployment rates, and higher living

standards, thus the beneficial effect of energy consumption should come as no

surprise. For this reason, the building sector is strongly reliant on the energy

industry to keep its operations running smoothly. As noted from the

construction industry, Table 5 also shows the effect of realised crude oil price

volatility on the electricity sub-sector.

Table 5: Quantile Regression Estimates (ROVX on ELECT)

Electricity Sub-sector
Τ OP ROVX EC INT FDI
Stress
0.05 0.0815 -0.1055 1.0843** -0.7164 0.0400
0.10 -0.1109 1.9656 1.5547*** -1.2656* 0.0150
0.15 -0.7880 -6.9583 1.6364*** -1.8243*** 0.2198
0.20 -1.3559*** -5.6443 1.4172*** -2.1837*** 0.4687***
0.25 -1.3244*** -6.1381 1.2174*** -2.1069*** 0.5246***
0.30 -1.2737*** -6.1101 1.1056*** -2.0811*** 0.5515***
0.35 -1.1798*** -6.7185 1.2664*** -2.2820*** 0.5184***
Normal
0.40 -1.1962*** -7.4064 1.2722*** -2.3521*** 0.5343***
0.45 -1.2548*** -8.8669** 1.4359*** -2.5041*** 0.5206***
0.50 -1.2681*** -6.4849 1.4576*** -2.4868*** 0.5144***
0.55 -1.3300*** -6.9157** 1.4645*** -2.5122*** 0.5298***
0.60 -1.3772*** -7.1541** 1.4132*** -2.4920*** 0.5535***
0.65 -1.3797*** -6.9978*** 1.4868*** -2.5557*** 0.5419***
Boom
0.70 -1.3680*** -7.2333*** 1.5234*** -2.5885*** 0.5342***
0.75 -1.4052*** -7.5121*** 1.4799*** -2.5851*** 0.5566***
0.80 -1.4380*** -8.3503*** 1.5907*** -2.6227*** 0.5362***
0.85 -1.3891*** -7.9962*** 1.4728*** -2.5248*** 0.5514***
0.90 -1.1655*** -6.6657** 1.2601*** -2.3081*** 0.5461***
0.95 -0.9081*** -3.3960* 1.0088*** -2.1240*** 0.5512***
Source: Author’s Construct (2023)
Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

79

Digitized by Sam Jonah Library

University of Cape Coast [Link]

From Table 5, there is a negative significant effect of ROVX on

electricity sub-sector at middle quantile (0.45, 0.55, 0.60 and 0.65) and upper

quantile (0.70-0.95) respectively representing normal and boom economic

situations. However, unlike the construction industry, the electricity sub-sector

is susceptible to negative shocks from ROVX at normal market situations in

addition to boom. That is, effect from the ROVX is crucial during normal and

boom market situations of the electricity industry. Conversely, the electricity

industry output is insulated against negative shocks from ROVX during

stressed conditions of the electricity industry output. For investors, it is better

to diversify or hedge against volatility in the electricity sub-sector by

observing the ROVX with other comparable investible assets, for instance, the

crude oil price.

Additionally, the control variables have significant effect on the

electricity sub-sector at most quantiles as found for the construction sector. As

a result, the electrical sector is negatively impacted by rising crude oil prices

in all three economic conditions (stress, normal, and boom – quantiles 0.20-

0.95). However, investors of the energy market and the electricity sub-sector

may find the markets attractive with the quest of offering diversification

benefits. This outcome is similar to that of the nexus between interest rate and

the electricity sub-sector (quantiles 0.10-0.95). On the other hand, the growth

of the electricity sub-sector is positively correlated with both the consumption

of energy (quantiles 0.05-0.95) and the inflow of direct foreign investment

(quantiles 0.20-0.95). Table 6 shows the results of the study's further

investigation into the effect of crude oil price volatility on the manufacturing

sub-sector.

80

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Table 6: Quantile Regression Estimates (ROVX on MFG)

Manufacturing Sub-sector
Τ OP ROVX EC INT FDI
Stress
0.05 -0.1805 -1.3202 1.7648*** -1.3133*** 0.0889
0.10 -0.3651 -1.5071 1.6406*** -1.3222*** 0.1688**
0.15 -0.5948** -4.9105 1.5962*** -1.4734*** 0.2576**
0.20 -0.8929*** -3.8318 1.6406*** -1.8620*** 0.3722***
0.25 -0.7961*** -3.4565 1.5424*** -1.7897*** 0.3786***
0.30 -0.7196*** -3.1689 1.5170*** -1.7564*** 0.3696***
0.35 -0.6377*** -3.0271 1.5405*** -1.7794*** 0.3527***
Normal
0.40 -0.6849*** -3.9943 1.5619*** -1.8428*** 0.3688***
0.45 -0.6987*** -4.5195* 1.6469*** -1.8850*** 0.3522***
0.50 -0.7030*** -4.8974** 1.7038*** -1.9204*** 0.3417***
0.55 -0.6975*** -4.4918** 1.7789*** -1.9533*** 0.3226***
0.60 -0.6856*** -4.1301** 1.7875*** -1.9589*** 0.3193***
0.65 -0.6704*** -4.2188*** 1.7939*** -1.9563*** 0.3145***
Boom
0.70 -0.6297*** -3.6708** 1.8029*** -1.9597*** 0.3054***
0.75 -0.6328*** -3.4601*** 1.7335*** -1.9260*** 0.3241***
0.80 -0.6125*** -3.3732*** 1.7039*** -1.8736*** 0.3226***
0.85 -0.5830*** -3.3605*** 1.7276*** -1.8721*** 0.3095***
0.90 -0.5663*** -3.1841*** 1.6368*** -1.7981*** 0.3255***
0.95 -0.3948*** -2.4492*** 1.6005*** -1.7295*** 0.2949***
Source: Author’s Construct (2023)

Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

From Table 6, ROVX has a negative influence on the manufacturing

sub-sector output for the middle quantile (0.45-0.65) and upper quantile (0.70-

0.95). Consequently, the influence is substantial. during normal and boom

economic conditions of the manufacturing sub-sector output. This corresponds

81

Digitized by Sam Jonah Library

University of Cape Coast [Link]

to the outcome on the electricity sub-sector where the sector output is

susceptible to negative shocks from ROVX at normal market situations in

addition to boom. Again, effect of the ROVX is crucial during normal and

boom market situations of the manufacturing industry. Conversely, there is

insignificant nexus at the lower quantiles (0.05 and 0.10). It can be ascertained

that the influence of ROVX during stressed conditions of the manufacturing

sub-sector output is not substantial. For investors, it is better to diversify or

hedge against volatility in the manufacturing sub-sector by observing the

ROVX with other comparable investible assets, for instance, the crude oil

price.

Also, it is clear from Table 6 that the control variables have a

significant influence on the output of the manufacturing sub-sector across the

majority of quantiles, just as was shown for the building and power industries.

An increase in the cost of crude oil has an adverse impact on industrial output

under all market situations, including stress (0.15-0.35) normal (040-0.65) and

boom (0.70-0.95) economic conditions. However, investors of the energy

market and the manufacturing industry may find the markets attractive with

the quest of offering diversification benefits. The correlation between interest

rates and the manufacturing sub-sector is analogous to this finding (0.05-0.95).

Meanwhile, the industrial sector does well when there is an increase in both

energy consumption (quantiles 0.05-0.95) and FDI (quantiles 0.10-0.95).

Table 7 provides a deeper dive into how realised crude oil volatility has

affected the mining and quarrying sub-sector.

82

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Table 7: Quantile Regression Estimates (ROVX on MQ)

Mining and Quarrying Sub-sector
Τ OP ROVX EC INT FDI
Stress
0.05 -0.1438 -0.6568 0.5940*** -0.1396 0.2460***
0.10 -0.2732 -1.7686 0.7655 -0.3059 0.2443***
0.15 -0.4767 -3.2958 1.0844 -0.7278 0.2516**
0.20 -1.1394*** -4.3880 2.0248*** -2.2172*** 0.3294**
0.25 -1.1272*** -5.4421 1.7852*** -2.0602*** 0.3877***
0.30 -0.8143*** -3.4737 1.9409*** -2.0838*** 0.2867***
0.35 -0.7157*** -3.3871 1.9607*** -2.1444*** 0.2747***
Normal
0.40 -0.7218*** -6.0440 2.3518*** -2.6200*** 0.2335***
0.45 -0.6593*** -6.0719 2.3553*** -2.6686*** 0.2311***
0.50 -0.6466*** -6.4525 2.3452*** -2.7092*** 0.2417***
0.55 -0.6938*** -6.9651* 2.3395*** -2.7043*** 0.2534***
0.60 -0.6987*** -7.0970** 2.2778*** -2.6700*** 0.2709***
0.65 -0.6626*** -5.6127** 2.2743*** -2.6883*** 0.2700***
Boom
0.70 -0.6289*** -5.6281*** 2.2930*** -2.6758*** 0.2566***
0.75 -0.4482*** -5.3095*** 2.3194*** -2.6750*** 0.2163***
0.80 -0.4363*** -5.3711*** 2.1901*** -2.6377*** 0.2533***
0.85 -0.3899*** -5.3097*** 2.2539*** -2.6626*** 0.2291***
0.90 -0.3430*** -5.8517*** 2.3465*** -2.7128*** 0.2001***
0.95 -0.3453*** -5.5080*** 2.1769*** -2.5725*** 0.2349***
Source: Author’s Construct (2023)

Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

83

Digitized by Sam Jonah Library

University of Cape Coast [Link]

From Table 7, ROVX has a negative influence on the mining sub-

sector output for middle quantile (0.55-0.65) and upper quantile (0.70-0.95)

correspondingly for normal and boom economic conditions of the mining sub-

sector output. The effect is significant during normal and boom economic

conditions. This corresponds to the outcome on the electricity sub-sector

where the sector output is susceptible to negative shocks from ROVX at

normal market situations in addition to boom. Again, effect of the ROVX is

crucial during normal and boom market situations of the mining sub-sector.

On the contrary, the mining sub-sector output is protected against significant

negative shocks from ROVX during stressed economic conditions of the

mining sub-sector output as found for electricity and manufacturing sub-

sector. For investors, it is better to diversify or hedge against volatility in the

mining industry by noticing the ROVX with other comparable investible

assets, for instance, the crude oil price.

From Table 7, as was found for the construction, electricity and

manufacturing sub-sectors, the control variables also have a considerable

effect on the mining sub-sector output at the majority of quantiles of the

mining sub-sector. Hence, increases in the price of crude oil have a

detrimental effect on the mining sub-sector output during times of stress

(quantiles 0.20-0.35), normal (quantiles 0.40-0.65), and boom (quantiles 0.70-

0.95) economic conditions of the mining sub-sector. However, investors of the

energy market and the mining industry may find the markets attractive with

the quest of offering diversification benefits. This result is comparable to the

relationship between interest rates and the mining sector (quantiles 0.20-0.95).

On the other hand, energy consumption (quantiles 0.05, 0.20-0.95) and foreign

84

Digitized by Sam Jonah Library

University of Cape Coast [Link]

direct investment (quantiles 0.05-0.95) are positively correlated with the

output of the mining sector. The influence of ROVX on water and sewerage

sub-sector output is further presented in Table 8.

Table 8: Quantile Regression Estimates (ROVX on WS)

Water and Sewerage Sub-sector
Τ OP ROVX EC INT FDI
Stress
0.05 -0.1234 1.5999 1.5354*** -1.3521*** 0.0079
0.10 -0.2071 0.5804 1.8395*** -1.6842*** -0.0189
0.15 -0.6921 -7.1092 1.6144*** -1.8905*** 0.1905
0.20 -1.3150*** -5.4117 1.4659*** -2.3203*** 0.4376**
0.25 -1.2510*** -6.2729 1.3506*** -2.2961*** 0.4711***
0.30 -1.2475*** -6.6241 1.2712*** -2.3359*** 0.5060***
0.35 -1.1489*** -7.1248 1.4380*** -2.5125*** 0.4658***
Normal
0.40 -1.0912*** -6.9703 1.4290*** -2.5203*** 0.4609***
0.45 -1.2095*** -8.1569* 1.3979*** -2.5427*** 0.5002***
0.50 -1.2525*** -6.6278* 1.4131*** -2.5444*** 0.5050***
0.55 -1.2737*** -5.6783** 1.4411*** -2.5906*** 0.5086***
0.60 -1.3246*** -6.2963** 1.4347*** -2.5819*** 0.5210***
0.65 -1.3279*** -6.4035*** 1.4752*** -2.6210*** 0.5163***
Boom
0.70 -1.3488*** -6.6066*** 1.4640*** -2.6208*** 0.5249***
0.75 -1.3821*** -6.8937*** 1.4320*** -2.6275*** 0.5443***
0.80 -1.4030*** -7.6637*** 1.5393*** -2.6412*** 0.5192***
0.85 -1.3874*** -7.4969*** 1.4456*** -2.5667*** 0.5364***
0.90 -1.1964*** -6.3528** 1.2547*** -2.3688*** 0.5330***
0.95 -0.9047*** -3.2899* 0.9868*** -2.1664*** 0.5343***
Source: Author’s Construct (2023)
Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

85

Digitized by Sam Jonah Library

University of Cape Coast [Link]

As presented in Table 8, ROVX has a negative influence on the water

and sewerage sub-sector output at middle quantile (0.45-0.65) and upper

quantile (0.70-0.95). The effect is significant during normal and boom market

conditions. This corresponds to the outcome of the electricity, manufacturing

and mining sub-sectors where the sub-sectors’ output is susceptible to negative

shocks from ROVX at normal market situations in addition to boom. Again,

the effect of the ROVX is crucial during normal and boom market situations

of the water and sewerage sub-sector output. On the contrary, the water and

sewerage sub-sector output is protected against significant negative shocks

from ROVX during stressed conditions of the water and sewerage sub-sector

output as found for electricity, manufacturing and mining sub-sectors.

The control variables also have a significant effect on the output of the

water and sewerage sector at the majority of quantiles of the water and

sewerage sub-sector output, as indicated in Table 8, as was discovered for the

construction, electricity, manufacturing, and mining sub-sectors. Hence, rises

in the price of crude oil have a negative effect on the output of the water and

sewerage sub-sector during stressful times (quantiles 0.20-0.35), normal

(quantiles 0.40-0.95), and boom (quantiles 0.65-0.95). It is advisable to use

caution when detecting the negative nexus at all quantiles because Kassim and

Isik (2020) thought that a lack of energy supply and inefficient energy

consumption posed a threat to industrial expansion. Investors in the energy

market and the water and sewerage sub-sector, however, can find these sectors

appealing due to the potential for diversification benefits.

After ascertaining the direct effect of ROVX on the five sub-sectors,

the study proceeds to decipher the vulnerability of the aggregated industrial

86

Digitized by Sam Jonah Library

University of Cape Coast [Link]

sector output to realised crude oil volatility. This is relevant for resource

management decisions because it will save time and resources to observe the

overall industry sector for shocks transmission from ROVX if the overall

sector output corresponds to the five sub-sectors. It goes to reason that it is

important to lay much emphasis on the overall industry sector output as the

point of call in the examination of the vulnerability of the industrial sector if it

strongly reflects the sub-sectors dynamics. This assertion is built on the theory

of financial and economic integration where markets do not operate in isolated

system as supported by extant literature (Asafo-Adjei, Adam, Arthur, Seidu &

Gyasi, 2022; Osei & Adam, 2020; Owusu Junior, Adam, Asafo-Adjei,

Boateng, Hamidu & Awotwe, 2021). Table 9 further details the effect of

ROVX on the industrial sector output.

Table 9: Quantile Regression Estimates (ROVX on ISO)

Industrial Sector Output
Τ OP ROVX EC INT FDI
Stress
0.05 -0.1678 -0.9784 1.1609*** -0.6348** 0.2063***
0.10 -0.3700 -1.9970 1.1045** -0.6794 0.2741***
0.15 -0.6084 -3.7062 1.4270** -1.1141* 0.2892**
0.20 -1.1028*** -4.6935 1.8360*** -2.0152*** 0.4106***
0.25 -1.0219*** -5.3889 1.6844*** -1.8634*** 0.4267***
0.30 -0.8691*** -3.7828 1.6833*** -1.8761*** 0.4029***
0.35 -0.6896*** -3.3872 1.7401*** -1.9223*** 0.3627***
Normal
0.40 -0.7398*** -5.9303 2.0758*** -2.3366*** 0.3365***
0.45 -0.7304*** -6.1595 2.1161*** -2.3313*** 0.3226***
0.50 -0.7425*** -6.2392** 2.0832*** -2.2899*** 0.3300***
0.55 -0.7716*** -5.4500** 1.9698*** -2.2382*** 0.3652***
0.60 -0.7304*** -5.5932*** 1.9656*** -2.2384*** 0.3602***
0.65 -0.7216*** -4.2822** 1.9216*** -2.2305*** 0.3718***
Boom
0.70 -0.7012*** -4.5606*** 1.8381*** -2.1758*** 0.3881***
0.75 -0.6150*** -3.7553*** 1.7548*** -2.1239*** 0.3918***
0.80 -0.5869*** -3.7896*** 1.7470*** -2.1255*** 0.3905***

87

Digitized by Sam Jonah Library

University of Cape Coast [Link]

0.85 -0.5906*** -3.9217*** 1.7378*** -2.1102*** 0.3930***

0.90 -0.5176*** -3.4465*** 1.6154*** -2.0323*** 0.4078***
0.95 -0.4215*** -3.0463*** 1.5825*** -1.9735*** 0.3920***
Source: Author’s Construct (2023)

Note: The variables OP, ROVX, EC, INT, and FDI represent Crude oil price,
Realised crude oil price volatility, Energy consumed, Interest rate and
Foreign Direct Investment respectively. It must be noted that τ signifies the
quantiles of the distribution. Also, ***, ** and * denote significance at 1%,
5% and 10% respectively.

From Table 9, ROVX has a negative significant effect on ISO at the

middle quantile (0.50-0.65) and upper quantile (0.70-0.95) respectively for

normal and boom economic conditions. This corresponds to the outcome on

the electricity, manufacturing, mining and water and sewerage sectors

representing four sectors out of five with greater susceptible to negative

shocks from ROVX at normal and boom economic conditions. Again, effect

of the ROVX can be found during normal and boom market situations of the

overall industry output. On the contrary, the overall industry output is guarded

against significant negative shocks from ROVX during stressed conditions of

the overall industry output as found for the five sub-sectors.

The control variables (OP, EC, INT and FDI) are seen to have

significant effects on ISO at varying quantiles. For instance, an increase in

crude oil prices has a detrimental effect on the output of the industry at stress

(quantiles 0.20-0.35), normal (quantiles 0.40-0.65), and boom (quantiles 0.70-

0.95) economic conditions. Yet, due to the potential for gains from

diversification, investors within the energy market as well asthe output of the

broader industry may find these areas intriguing. This result is comparable to

the connection between interest rates and the output of the entire industry

88

Digitized by Sam Jonah Library

University of Cape Coast [Link]

(quantiles 0.05-0.95). The output of the total industry, on the other hand, is

positively correlated with both energy consumption (quantiles 0.05-0.95) and

foreign direct investment (quantiles 0.05-0.95).

Discussion of the asymmetric relationship between realised crude oil price

volatility and industrial sector output

In response to the first research objective, realised crude oil price

volatility has a negative influence on the five sub-sectors and overall industry

output. However, the effect differs marginally across the sub-sectors. For

instance, the electricity, manufacturing, mining and quarrying and water and

sewerage sectors representing four sub-sectors out of five had greater

susceptibility to negative shocks from realised crude oil price volatility at

normal and boom economic conditions. This implies that industrial sub-sectors

that rely heavily on oil may experience lower economic growth during normal

and booming economic conditions of the industrial sector output. This can

result in reduced investment levels and elevated unemployment rates,

potentially triggering a cascading impact on the overall economy.

The study indicates that escalating oil prices result in heightened

volatility in crude oil prices, negatively impacting the industrial sector's output

due to higher production costsTherefore, the production costs of the industrial

sector which depends heavily on oil would increase which lowers the sector’s

output. Additionally, oil price volatility can also affect consumer and business

confidence, which can lead to changes in consumption and investment

behavior. When oil prices are volatile, businesses may be less confident about

the future, leading to lower levels of consumption and investment, thereby

dwindling industrial sector output.

89

Digitized by Sam Jonah Library

University of Cape Coast [Link]

The detrimental impact of sudden crude oil price fluctuations on the

industrial sector output is supported by prior studies such as Ahmed et al.

(2017) using the VAR approach, and Zhang et al. (2022) in China.

Conversely, Iganiga et al. (2021) found a positive effect of crude oil price

shocks on construction sub-sector and the overall industrial sector but negative

effect on the manufacturing sub-sector in Nigeria when the ARDL and

NARDL were employed.

Accordingly, the effect of the realised crude oil price volatility can be

found during normal and boom economic situations of the industrial sector

output such as in the electricity, manufacturing, mining and water and

sewerage sub-sectors. The effect of ROVX on the industrial sector output was

seen to be insignificant mostly during stressed conditions. This is possible

because when the industrial sector experiences stressful conditions,

consumption of refined crude oil might not increase which mitigates the

adverse effect of crude oil price volatility.

The results of this study hold practical significance for investors in the

Ghanaian industrial sectors. It suggests that investors in the electricity,

manufacturing, mining and quarrying, and water and sewerage sub-sectors

may face greater risks from volatile crude oil prices during normal and boom

economic conditions. The study's theoretical implications suggest that the

effect of crude oil price volatility on the industrial sectors is not uniform

across quantiles. This finding aligns with earlier research (Dawar, Dutta,

Bouri, and Saeed, 2021; Zhu, Guo, You, and Xu, 2016; Hamdi, Aloui,

Alqahtani, and Tiwari, 2019), demonstrating that the impact of fluctuations in

crude oil prices on the economy can differ based on the prevailing oil price

90

Digitized by Sam Jonah Library

University of Cape Coast [Link]

levels. This suggests that the relationship between crude oil price volatility is

asymmetric thereby violating the symmetric/linear relationship growth theory.

The Conditional Causality of Realised Crude Oil Price Volatility to

Industrial Sector Output
Conditional Causality Model Results

This section presents the outcome on the conditional causality in mean

as a way of confirming causality from ROVX to the industrial sector output.

Since the approach is bivariate, causality can only be ascertained from ROVX

to the industrial sector output with no control variables considered.

Explanations of the quantiles from this approach is similar to the quantile

regression technique only that the quantile regression does not depict

causality. Similarly, the lower quantile ranges from 0.05-0.35 (stressful), the

middle quantile from 0.40-0.65 (normal), and the higher quantile from 0.70-

0.95 (boom) are used to illustrate economic conditions of industrial sector

output (see, Adebayo et al., 2022; Demir et al., 2020).

In the conditional causality in quantile results, the test statistics

correspond to the vertical axis with the quantiles presented on the horizontal

axis in each plot. It is observable from the 5% significance level in connection

with the critical value of 1.96 as illustrated by the horizontal solid line in the

causality results. To illustrate significance at specific quantiles, the yellow

curve should be above the 1.96 solid line before the null hypothesis (a change

in ROVX does not cause a change in industrial sector output) can be rejected.

Figure 8 presents the conditional causality in quantile between ROVX and the

construction sub-sector output.

91

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 8: Quantile Causality in Mean from Realised Crude Oil Volatility to

the Construction Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables CONS, and ROVX represent Construction sub-sector and
Realised crude oil price volatility.
From Figure 8, it is evident that fluctuations in realised crude oil price

volatility (ROVX) effect the output of the construction sub-sector at economic

stress, as depicted by the lower quantiles (0.05-0.35). This highlights the

conditional causality between ROVX and the construction sub-sector output.

The construction sub-sector is one of the critical economic sectors that can be

directly impacted by fluctuations in oil prices. For instance, changes in oil and

gas prices can affect construction costs, transportation expenses, and demand

for construction projects, which ultimately impact the output of the

construction sub-sector. Figure 9 presents the conditional causality in quantile

between ROVX and the electricity sub-sector output.

92

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 9: Quantile Causality in Mean from Realised Crude Oil Price Volatility
to the Electricity Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ELECT, and ROVX represent Electricity sub-sector and
Realised crude oil price volatility.

From Figure 9, it is evident that there is no causal relationship between

realised crude oil price volatility (ROVX) and the output of the electricity sub-

sector, regardless of the economic conditions of the electricity sub-sector

output. This indicates that ROVX does not drive the electricity sub-sector

output at varying quantiles, all other things held constant. This can be traced

from the yellow curve below the 1.96 solid line which indicates that the null

hypothesis (a change in ROVX does not cause a change in industrial sector

output) is not rejected. The electricity generation in Ghana is sourced from a

variety of energy resources, encompassing hydroelectric power, natural gas,

and renewable energy. While crude oil might play a role in the energy sector,

its impact may be diluted by the presence of other dominant energy sources.

The results from Figure 9 suggest that the electricity sub-sector may

not be as vulnerable to causality from crude oil price volatility as other

economic sectors. Figure 10 presents the conditional causality in quantile

between ROVX and the manufacturing sub-sector output.

93

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 10: Quantile Causality in Mean from Realised Crude Oil Price
Volatility to the Manufacturing Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables MFG, and ROVX represent Manufacturing sub-sector and
Realised crude oil price volatility.

Figure 10 demonstrates that realised crude oil price volatility (ROVX)

has a causal relationship with the output of the manufacturing sub-sector at

economic stress, as represented by the lower quantiles (0.2-0.35). This implies

that ROVX has the potential to cause the manufacturing sub-sector's output at

only the lower quantile, all other things held constant. The manufacturing sub-

sector's production is a pivotal component of the economy, and it is vulnerable

to changes in the volatility of crude oil prices. The manufacturing sector relies

heavily on petroleum products as raw materials and energy sources, making it

vulnerable to changes in oil and gas prices. Figure 11 presents the conditional

causality in quantile between ROVX and the mining and quarrying sub-sector

output.

94

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 11: Quantile Causality in Mean from Realised Crude Oil Price
Volatility to the Mining and Quarrying Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables MQ, and ROVX represent Mining and quarrying sub-
sector and Realised crude oil price volatility.

Figure 11 indicates that there is a causal relationship between realised

crude oil price volatility (ROVX) and the output of the mining and quarrying

sub-sector from the yellow curve above the 1.96 critical solid line. The

significant causality can be found at the lower (0.1-0.2) and middle quantiles

(0.55-0.6) representing stressed and normal economic conditions. This implies

that ROVX can drive the output of the mining and quarrying sub-sector at

varying quantiles, holding all other things constant. The fact that this causality

is most pronounced in the lower quantiles, representing stressed economic

conditions, suggests that the mining and quarrying sub-sector is particularly

sensitive to changes in crude oil prices during times of economic uncertainty.

This sensitivity can be linked to a multitude of factors, including increased

production costs and reduced demand for raw materials in turbulent economic

environments (Humssi, Petrovskaya, and Abueva, 2022). Figure 12 shows the

conditional causality in quantile between ROVX and the water and sewerage

sub-sector output.

95

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 12: Quantile Causality in Mean from Realised Crude Oil Price
Volatility to the Water and Sewerage Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables WS, and ROVX represent Water and Sewerage sub-sector
and Realised crude oil price volatility.

From Figure 12, it is clear that there exists a causal connection

between the observed fluctuations in realized crude oil price volatility

(ROVX) and the output of the Water and Sewerage sub-sector. The significant

causal nexus can be observed from the yellow curve above the 1.96 critical

solid line in Figure 12. This relationship is observed at the middle quantiles

(0.3-0.55), representing stressed to normal economic conditions. This implies

that ROVX can drive the Water and Sewerage sub-sector output at varying

quantiles, holding all other things constant.

Hence, variations in oil price volatility can exert substantial impacts on

the Water and Sewerage sub-sector's output. In times of elevated oil prices, the

Water and Sewerage sub-sector can experience increased costs for energy and

raw materials, leading to reduced profitability. Conversely, low oil prices can

lead to decreased demand for water treatment services, resulting in lower

revenues for the sector. Figure 13 shows the conditional causality in quantile

between ROVX and the industrial output.

96

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 13: Quantile Causality in Mean from Realised Crude Oil Price
Volatility to Industrial Sector Output
Source: Author’s Construct (2023)
Note: The variables ISO, and ROVX represent Industrial sector output and
Realised crude oil price volatility.

From Figure 13, it is evident that the realised crude oil price volatility

(ROVX) has a causal relationship with the output of the Industrial sector,

specifically at the lower quantile (0.1-0.2) representing stressful economic

conditions. This implies that ROVX can drive the Industrial sector output at

only the lower quantile, holding all other things constant. The Industrial sector

comprises various sub-sectors that depend on oil and gas products as inputs for

production and operations. These sub-sectors include manufacturing, mining

and quarrying, and construction, among others. Hence, fluctuations in oil

prices can have a significant effect on the Industrial sector's output. In times of

elevated oil prices, the Industrial sector can experience increased costs for

energy and raw materials, leading to reduced profitability. Conversely, low oil

prices can lead to decreased demand for the Industrial sector's products and

services, resulting in lower revenues for the sector.

Discussion of the conditional causality of realised crude oil price volatility

to industrial sector output

97

Digitized by Sam Jonah Library

University of Cape Coast [Link]

The findings presented on the conditional causality between ROVX

and industrial sector output showed that ROVX drives most of the industrial

sector output at lower and middle quantiles representing market stress and

normal market conditions, respectively. The outcome implies that crude oil

price volatility increases or decreases production costs, investors’ confidence,

and demand of oil products, among others, which determines output of the

industrial sector. The pattern of causality to the output of the industrial sector

is specific during stress conditions of the construction, manufacturing, mining

and industrial sector output, but at normal economic conditions of the water

and sewerage sector and manufacturing sector output. However, the

insignificant causality between crude oil price volatility and industrial sector

output at economic boom signifies that, variations in crude oil price volatility

does not drive the industrial sector output. Hence, when the industrial sector is

highly performing, the sector becomes less vulnerable to crude oil price

volatility.

Considering the causal effect of ROVX on output of the industrial

sector, outcome from this study corroborates the findings of Al-Sasi et al.

(2017) in the setting of United Arab Emirates as well as the study of Akalpler

et al. (2018) in Nigeria. Causality from ROVX to industrial macroeconomic

fundamental of Ghana also supports the findings of Archer et al. (2022) on the

vulnerability of exchange rate to crude oil price in Ghana. In addition to this,

Asafo-Adjei, Adam and Darkwa (2021) discovered a substantial correlation

between crude oil price and stock returns in Ghana. This result is consistent

with the findings of Hau et al. (2020), who utilized crude oil price volatility

and agricultural commodities in the context of China.

98

Digitized by Sam Jonah Library

University of Cape Coast [Link]

The Responsiveness of The Industrial Sector Output to Realised Crude

Oil Price Volatility Across Time and Frequency

Time Frequency Connectedness Model Results

In response to the time and frequency connectedness between ROVX

and industrial sector output, the bi-wavelet approach is employed. This

approach is capable of extracting time and frequency dimensions of the nexus.

The time dimension shows the calendar time whereas the frequency

perspective divulges the timeframes for investment which encompass short-

term, medium-term, and long-term horizons. The frequencies are denoted by

scales represented on the vertical axis on the extreme left. The calendar times

are shown on horizontal axis on top of the bi-wavelet plot. Following the

studies of Asafo-Adjei, Boateng, Isshaq, Idun, Owusu Junior and Adam

(2021), and Boateng, Asafo-Adjei, Addison, Quaicoe, Yusuf and Adam,

(2022), the study considers the short-term as from scales 0-8, between scales 8

and 32 as medium-term and beyond scale 32 as long-term.

The study follows discussions by extant literature (Boateng, Asafo-

Adjei, Addison, Quaicoe, Yusuf & Adam, 2022; Idun et al., 2022; Nkrumah-

Boadu et al., 2022; Singh, Bansal & Bhardwaj, 2022) to help explain the

results. In graphical terms, time series that are in-phase are denoted by arrows

heading to the right (left) (out of phase). Arrows from the bi-wavelet plots

geared toward left denote negative nexus whereas arrows inclined to the right

show positive co-movements. To decipher the pattern of leading and lagging

variables, diagonal arrows are used.

Hence, a left arrow pointing downward or a right arrow pointing

upward suggests that the second variable lags (or the first variable leads) the

first by ⁄ , whereas left arrow pointing upward or a right arrow pointing

99

Digitized by Sam Jonah Library

University of Cape Coast [Link]

downward indicates that the first time series lags the second by ⁄ . In this

case the first time series is ROVX whereas the second time series is the

industrial sector output. Significant co-movements are denoted by regions in

red (warm) colour whereas insignificant co-movements are shown in blue

(cool) colour. This is pictorially highlighted by the colour bar and the

corresponding magnitude of the nexus located on the right-side of each bi-

wavelet plot. The time and frequency connectedness between ROVX and the

construction sub-sector output is shown in Figure 14.

Figure 14: Co-movements between Realised Crude Oil Volatility and

Construction Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and CONS represent Realised crude oil price
volatility and Construction sub-sector output respectively

Figure 14 depicts the relationship between ROVX and the

Construction sub-sector output over time and frequency, highlighting the

extent of mutual dependence between the variables. The results from Figure

14 depicted by the warm (red) colours indicate significant co-movements

between ROVX and Construction sub-sector output in the short-term (scales

2-6) during the periods of 2003-2008 and 2015-2016. Additionally, in the

medium-term (scales 8-32) and early sections of the long-term (scales 32-40),

100

Digitized by Sam Jonah Library

University of Cape Coast [Link]

strong co-movement is observed between 2015 and 2019. The correlation

between ROVX and Construction sub-sector output shows a mix of positive

(between 2016 and 2017 – scales 8-32) and negative (in 2015 – scales 2-6)

nexus as indicated by the right and left pointing arrows respectively.

Notably, the negative co-movement between ROVX and Construction

sector output during the short-term (scales 2-6) in 2015, has ROVX lagging. In

contrast, in the medium-term (scales 8-32) between 2016 and 2017, ROVX

drives Construction industry output. It is crucial to observe the dynamics of

the Construction sub-sector output and its relationship with ROVX, especially

during the short-term (scales 2-6), as they could be susceptible to shocks from

ROVX. Figure 15 presents the time and frequency nexus between ROVX and

the electricity sub-sector output.

Figure 15: Co-movements between Realised Crude Oil Volatility and

Electricity Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and ELECT represent Realised crude oil price
volatility and Electricity sub-sector output respectively

Figure 15 highlights the relationship between ROVX and the

Electricity sub-sector output. The short-term (scales 0-8) co-movements

between ROVX and Electricity sub-sector output from 2003 to 2009 and from
101

Digitized by Sam Jonah Library

University of Cape Coast [Link]

2015 to 2017 are particularly strong. Additionally, the medium (scales 8-32) to

long-term (32-50) interconnectedness between 2015 and 2019 is also strong as

depicted by the red (warm) colour. Interestingly, both positive (right arrows)

and negative (left arrows) relationships are observed between ROVX and

Electricity sub-sector output.

In particular, the positive (right arrows) interconnectedness from

Figure 15 between ROVX and Electricity sub-sector output from 2015 to 2016

in the long-term (scales 32-50), with ROVX lagging, is notable. This is

denoted by the right-pointing arrows downward. However, a different

outcome is found for the interconnectedness between ROVX and Electricity

sub-sector output during the 2008 Global Financial Crisis. Specifically,

negative (left arrows) co-movements exist between ROVX and Electricity

sub-sector output, with ROVX leading the relationship in the short-term

(scales 0-8) through to the medium-term (scales 8-14) between 2007 and 2010.

ROVX leads because of the left-pointing arrows downwards. Figure 16 shows

the time and frequency co-movements between ROVX and the manufacturing

sub-sector output.

102

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 16: Co-movements between Realised Crude Oil Price Volatility and
Manufacturing Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and MFG represent Realised crude oil price
volatility and Manufacturing sub-sector output respectively

Figure 16 reveals the interconnectedness between ROVX and

Manufacturing sub-sector output across time and frequency. The short-term

(scales 0-6) analysis between 2003 and 2008, as well as between 2014 and

2017, reveals strong co-movements (shown in red colour) between ROVX and

Manufacturing sub-sector output. In the medium-term (scales 12-16) and early

sections of the long-term (scales 16-50), the interconnectedness is strongly

depicted by the red colour, primarily between 2015 and 2019.

Specifically, there is neither negative nor positive connectedness

between ROVX and Manufacturing sector output between 2015 and 2016 in

the short-, medium-, and long-terms. On the other hand, ROVX drives (from

the left-pointing arrows downwards) Manufacturing sub-sector output in the

short-term (scales 2-6) between 2007 and 2009. The strong connectedness

between 2007 and 2009 can be partially attributed to the 2008 Global

Financial Crisis, which had a significant effect on the global economy and the

oil and gas industry. Figure 17 illustrates the time and frequency co-

movements between ROVX and the mining and quarrying sub-sector output.

103

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Figure 17: Co-movements between Realised Crude Oil Price Volatility and
Mining and Quarrying Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and ELECT represent Realised crude oil price
volatility and Mining and quarrying sub-sector output respectively

Figure 17 highlights the interconnectedness between ROVX and the

Mining and Quarrying sub-sector. The short-term (scales 0-6) and medium-

term (scales 12-16) co-movements show that there is a strong relationship

(shown in red colour) between ROVX and the Mining and Quarrying sub-

sector output, particularly during the periods between 2003 and 2008, and

between 2015 and 2016.

Interestingly, there is neither negative nor positive connectedness

between ROVX and Mining and Quarrying sector output between 2015 and

2016 in the short-term, medium-term, and long-terms. This outcome is similar

to the manufacturing sector output. On the other hand, ROVX drives Mining

and Quarrying sub-sector output in the short-term (scales 2-6) between 2007

and 2009. The strong interconnectedness observed between 2007 and 2009

can partly be attributed to the 2008 Global Financial Crisis, which had a

significant effect on the global economy and the oil and gas industry. Figure

104

Digitized by Sam Jonah Library

University of Cape Coast [Link]

18 presents the time and frequency co-movements between ROVX and the

water and sewerage sub-sector output.

Figure 18: Co-movements between Realised Crude Oil Price Volatility and
Water and Sewerage Sub-sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and WS represent Realised crude oil price
volatility and Water and sewerage sub-sector output respectively

Figure 18 shows the interconnectedness between ROVX and Water

and Sewerage sub-sector output over time and frequency. It is observed that

there are strong co-movements (from the red colour) between ROVX and

Water and Sewerage sub-sector output in the short-term (scales 0-8) between

2003 and 2010, as well as between 2015 and 2016. Also, in the medium-term

(scales 8-16) strong co-movements can be found between 2007 and 2010, as

well as the early sections of the long-term, the interconnectedness is strong

between 2014 and 2019. The relationship between ROVX and Water and

Sewerage sub-sector output is a mix of both positive and negative.

Notably, there is negative (left arrows) co-movements between ROVX

and Water and Sewerage sector output between 2007 and 2010 in the short-

term (scales 0-8) and medium-term (scales 8-16), with ROVX leading. ROVX

leads at this point because of the left-pointing arrows downward. On the other

105

Digitized by Sam Jonah Library

University of Cape Coast [Link]

hand, ROVX lags (from the right-pointing arrows downward) Water and

Sewerage sub-sector output in the long-term between 2014 and 2019, which is

similar to the Electricity sub-sector. It is suggested that the strong

interconnectedness between 2007 and 2010 can partly be attributed to the

2008 Global Financial Crisis. In conclusion, the output of the Water and

Sewerage sub-sector is closely related to ROVX, with a mix of positive and

negative relationships. Figure 19 presents the time and frequency nexus

between ROVX and the industrial output.

Figure 19: Co-movements between Realised Crude Oil Price Volatility and
Industrial Sector Output
Source: Author’s Construct (2023)
Note: The variables ROVX and ISO represent Realised crude oil price
volatility and Industrial sector output respectively

Figure 19 shows the relationship between ROVX and the industrial

sector output. It indicates strong negative (left arrows) co-movements (from

the red colour) in the short-term (scales 2-6) between 2003 and 2009, and

between 2015 and 2016, which is consistent with the patterns observed in the

sub-sectors. Similarly, there is strong co-movement between the variables in

the medium-term (scales 14-16) and long-term (scales 32-50), between 2014

106

Digitized by Sam Jonah Library

University of Cape Coast [Link]

and 2019. During this period, there exist right-pointing arrows which indicates

positive nexus between ROVX and the industrial sector output. It is important

to note that the industrial sector encompasses a wide range of activities,

including manufacturing, construction, mining, and quarrying, among others.

Therefore, the pattern of co-movements observed in the industrial

sector is similar to that of the sub-sectors, reflecting the interconnectedness

between the ROVX and the broader economy. Generally, as revealed in the

medium-term, and long-terms co-movements are strong for the output of the

sub-sectors spanning important crises such as the BREXIT in 2016, banking

sector financial crisis in 2017 that occurred in Ghana as well as the COVID-19

pandemic. This highlights the effects between the realised crude oil volatility

and the industrial sector output. It can therefore be said that

interconnectedness between realised crude oil volatility and the industrial

sector are susceptible to crises which should be observed with caution by

resource managers.

Discussion on the responsiveness of the industrial sector output to realised

crude oil price volatility across time and frequency

The study found significant short-term correlations between the

volatility of crude oil prices and industry sector output between 2003 and 2008

as well as between 2015 and 2016. Beyond 2014, the interconnectedness was

significant in the medium-term and the early parts of the long-term. Between

2007 and 2010, there were specifically negative co-movements between

realised crude oil volatility and the output of the electricity sub-sector as well

as between realised crude oil price volatility and the output of the water and

sewerage industry, with realised crude oil price volatility leading the

relationship. The 2008 Global Financial Crisis can be partially attributed for

107

Digitized by Sam Jonah Library

University of Cape Coast [Link]

the significant co-movements between 2007 and 2010. The time and

frequency connectedness between ROVX and industrial sector output provides

that the nexus is adaptive and heterogeneous which corroborates the results by

Mo et al. (2019), Riaz et al. (2016) in Pakistan, Yu et al. (2022), and Zhang et

al. (2022) in China.

It should be emphasized that, prior to 2014, the co-movements were

primarily negative, with the productivity of the industrial sector being caused

by the realised volatility of crude oil. The output of the industrial sector shows

a similar pattern of causation with the realised volatility of crude oil price at

particular intervals and frequencies. As seen in the medium- and long-term

interconnectedness, the industrial sector has performed successfully despite

significant crises like the BREXIT in 2016, the Ghanaian banking sector

financial crisis in 2017, and the COVID-19 pandemic. This highlights the

effects between the realised crude oil volatility and the industrial sector output.

It can be established that co-movements between realised crude oil volatility

and the industrial sector are susceptible to crises which should be observed

with caution by resource managers.

The interconnectedness between the variables is not strong across all

times and frequencies but they are rather time and frequency specific. This

addresses the heterogenous nature and adaptive behaviors of the markets in

responds to the heterogeneous market hypothesis (Müller et al., 1997) and the

adaptive market hypothesis (Lo, 2004). This is in consonance with the

outcomes revealed by extant literature (Asafo-Adjei et al., 2021; Boateng,

Asafo-Adjei, Addison, Quaicoe, Yusuf & Adam, 2022; Boateng et al., 2021;

Li, Huang & Failler, 2022). The weaker co-movements shown in blue colour

108

Digitized by Sam Jonah Library

University of Cape Coast [Link]

(cold colour) are areas where crude oil volatility and the industrial sector

output are less connected. Accordingly, the industrial sector output is insulated

against adverse shocks from the reaslised crude oil volatility.

The results of this research carry significant implications for existing

policies in the Ghanaian context, particularly those related to petroleum

subsidies. The study highlights the negative effect of realised crude oil

volatility on the output of the industrial sector, particularly the electricity,

manufacturing, mining, and water and sewerage sub-sectors. This implies that

policies that aim to reduce subsidies on petroleum products may cause an

increase in production costs and a decline in the output of these sub-sectors.

For instance, the reduction in subsidies by the Ghanaian government

on petroleum products in periods such as 2013 and 2015 led to an eventual

increase in production cost of firms which dwindled output. The government

of Ghana in 2015 set aside US$12.5m for subsides, compared with the

US$150m in 2014 (Economist Intelligence, 2015). This reduction in subsidies

may have had a negative effect on the output of the industrial sector,

particularly the sub-sectors identified in this study.

Chapter Summary

The study quantitatively examined the effect of realised crude oil price

volatility on the industrial output of Ghana. For this reason, the quantile

regression, nonlinear conditional causality in quantile and bi-wavelet

approaches were utilised as the study’s estimation techniques. It was revealed

that realised crude oil price volatility has an asymmetric, causal influence and

time-frequency effect on industrial sector output. Categorically, the nexus

109

Digitized by Sam Jonah Library

University of Cape Coast [Link]

between realised crude oil price volatility and the industrial sector output is

asymmetric, heterogenous and adaptive.

CHAPTER FIVE

SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

Introduction

This research investigates the effect of crude oil price volatility on

industrial sector output in Ghana. Results, interpretations, conclusions,

recommendation and suggestions for further study are summed up and

discussed in this chapter. The empirical analysis looked at the asymmetrical

110

Digitized by Sam Jonah Library

University of Cape Coast [Link]

connection between crude oil price volatility and the industrial sector, and the

causality from crude oil price volatility to quantiles of the industrial sector

output. The study also evaluated how the output of the industrial sector is

related to the volatility of the price of crude oil in terms of both time and

frequency.

Using monthly data from January 2001 through December 2020,

quantile regression, conditional causality and bi-wavelet techniques were

applied to estimate the responsiveness of the industrial sector output to

realised crude oil price volatility. The variables for the study include, crude oil

price volatility, manufacturing subsector, mining and quarrying subsector,

water and sewage subsector, construction subsector, electricity subsector and

industrial sector outputs, respectively. Other control variables included were

energy consumption, foreign direct investment, and interest rate.

Summary of Findings

The research discovered from the first research objective that realised

crude oil price volatility has a significant adverse effect on the industrial sector

and its sub-sectors, particularly during normal and boom economic conditions.

The electricity, manufacturing, mining, and water and sewerage sectors were

found to be more susceptible to negative shocks from crude oil price volatility.

The study underscores the importance of the asymmetric relationships

between crude oil prices and industrial sector output, which, imperatively

makes more effective resource management and policy decisions in the

Ghanaian context and beyond.

Furthermore, the second objective looked at the causal relationship

between crude oil price volatility and industrial output. The study revealed that

111

Digitized by Sam Jonah Library

University of Cape Coast [Link]

the output of the sub-sectors of the industrial sector in Ghana is significantly

affected by ROVX, especially during periods of market stress and normal

market conditions. At such times, the effect of ROVX on construction,

manufacturing, mining, and overall industrial sector output is significant.

Additionally, the water and sewerage sub-sector output and the manufacturing

sub-sector output are susceptible to shocks from the ROVX during normal

market conditions.

Finally, the third objective analysed the responsiveness of industrial

output to crude oil price volatility across time and frequency. The study found

a notable correlation between the volatility of crude oil prices and the output

of the industrial sector in Ghana. Prior to 2014, the co-movements were

mostly negative, with realised crude oil price volatility causing a decline in

industrial sector output. However, beyond 2014, the interconnectedness was

significant over an extended period, both medium term and long term,

indicating the industrial sector's resilience to crises like the COVID-19

pandemic and the Ghanaian banking sector financial crisis in 2017. The

negative interconnectedness between realised crude oil price volatility and the

electricity and water and sewerage industries from 2007 to 2010, with crude

oil volatility leading the relationship, can be partially ascribed to the 2008

Global Financial Crisis.

Conclusions

In the first place, the instability of oil prices has a detrimental

consequence on the output of the electricity sub-sector, as well as the

manufacturing sub-sector, the mining and quarrying sub-sector, both in the

normal and boom economic conditions. At the period of heightened market

112

Digitized by Sam Jonah Library

University of Cape Coast [Link]

stress, however, these subsectors are shielded from the potentially damaging

effects of volatility in the price of crude oil. However, volatility in the price of

crude oil has negative effect on the construction sub-sector only during

economic conditions characterized by a boom state; and during economic

conditions characterized by a stressful situation, the subsector is shielded from

the price shock. The volatility of the price of crude oil also has a negative

effect on the overall output of the industrial sector.

From the second research objective, the findings highlight the

conditional nexus between the ROVX and the industrial sector output.

Causality is prominent during stress conditions of the construction,

manufacturing, mining and industrial sector output, but at normal economic

conditions of the water and sewerage sector and manufacturing sector output.

However, when the industrial sector is highly performing, the sector is less

susceptible to fluctuations in crude oil price instability.

Finally, this research also analysed the time and frequency

connectedness of industrial output and crude oil price volatility. An intense

adverse correlation in the volatility of crude oil prices and construction,

electricity, water and sewerage, manufacturing and mining and quarrying

sectors was obtained. The construction, electricity, water and sewerage,

manufacturing, and mining and quarrying sectors are highly sensitive to shifts

in oil price volatility. As crude oil price volatility increases, these sectors tend

to experience decreased industrial output, which suggests that these industries

are vulnerable to fluctuations in crude oil prices. It can be concluded that

fluctuations in crude oil prices can exert substantial spillover effects on the

113

Digitized by Sam Jonah Library

University of Cape Coast [Link]

industrial output, indicating that they are closely linked to the dynamics of the

oil market.

Recommendations

Considering the results obtained from the first research objective, it is

imperative for policymakers and resource managers in the industrial sector to

develop robust contingency plans that account for different economic

scenarios. These plans should include strategies to maintain stable output

during periods of crude oil price volatility, including market stress and normal

market conditions. Furthermore, policymakers should encourage industrial

producers to enter into forward contracts or engage in hedging strategies to

lock in stable prices for their energy needs. This can be supported through

partnerships with financial institutions and commodity markets, allowing

industries to mitigate the risk of sudden price spikes which could otherwise

lead to higher costs and lower profits.

Secondly, policymakers should prioritize the establishment of strategic

oil reserves to address supply-side shocks resulting from Ghana's dependence

on crude oil imports. These reserves would serve as a buffer against sudden

disruptions in global oil supply, reducing supply risks and stabilizing the

country's reaction to global oil price variations. Simultaneously, the

government should promote and incentivize industries to diversify their

energy sources beyond fossil fuels. Encouraging the adoption of renewable

energy, energy efficiency measures, and alternative fuels can reduce

dependence on volatile oil prices.

In light of the third research's findings, the government should consider

implementing petroleum policies to mitigate the negative effects of crude oil

114

Digitized by Sam Jonah Library

University of Cape Coast [Link]

price volatility on the industrial sector. This includes establishing a Price

Stabilization Fund which allows the government to save windfall gains in

times of elevated crude oil prices and release these funds to offset rising costs

when prices fall. This helps stabilize domestic petroleum product prices,

ensuring that industrial consumers are shielded from sudden price fluctuations.

The government could establish upper and lower price bands for fuel to help

prevent extreme price fluctuations. When prices approach the upper band, the

government can temporarily reduce taxes to stabilize prices, while it can

increase taxes when prices fall towards the lower band. This approach

maintains price stability for the industrial sector. Also, the government can

consider targeted fuel subsidies that focus on protecting vulnerable industries

within the industrial sector. By providing subsidies to specific industries

facing significant cost increases due to oil price volatility, the government can

minimize the adverse effects on industrial output while managing fiscal

resources more efficiently.

To conclude, the government should establish a dedicated unit within

the Ministry of Energy and Petroleum tasked with monitoring and analyzing

crude oil price movements to effectively track and respond to oil price swings

and their influence on industrial production. This unit should regularly assess

the potential impact of price changes on the industrial sector and recommend

timely policy adjustments. Additionally, the government should enhance

transparency and information sharing with stakeholders, including industry

players and the public, to ensure informed decision-making and promote

confidence in the management of oil-related challenges.

115

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Suggestions for Further Studies

Future research on the empirical examination of the impact of crude oil

price volatility on the Ghanaian context should expand its scope beyond the

industrial sector to encompass various sectors such as finance, digital

technology, healthcare, tourism, and education. It should also consider the

forward-looking nature of implied crude oil price uncertainty, derived from

options markets, to better understand how businesses and investors perceive

and react to potential oil price fluctuations. Comparative studies with

neighbouring countries, examination of policy implications, long-term effects,

regional variations, and the inclusion of additional macroeconomic indicators

are also recommended. Furthermore, researchers can explore the

environmental and social of variations in oil price levels on Ghana's sectors to

offer a more thorough insight into multifaceted impacts.

116

Digitized by Sam Jonah Library

University of Cape Coast [Link]

REFERENCES

Abdulkareem, A., & Abdulkareem, K. A. (2016). Analysing oil price-

macroeconomic volatility in Nigeria. CBN Journal of Applied Statistics

(JAS), 7(1), 1.

Abokyi, E., Appiah-Konadu, P., Sikayena, I., & Oteng-Abayie, E. F. (2018).

Consumption of electricity and industrial growth in the case of

Ghana. Journal of Energy, 2018.

Ackah, C., Adjasi, C., & Turkson, F. (2016). Scoping Study on the Evolution

of Industry in Ghana, a Learning to Compete Working Paper no. 8.

Adams, Z.; Collot, S.; Kartsakli, M. (2020). Have commodities become a

financial asset? Evidence from ten years of Financialization. Energy

Econ, 89, 104769

Adebayo, T. S., Rjoub, H., Akinsola, G. D., & Oladipupo, S. D. (2022). The

asymmetric effects of renewable energy consumption and trade

openness on carbon emissions in Sweden: new evidence from quantile-

on-quantile regression approach. Environmental Science and Pollution

Research, 29(2), 1875-1886.

Adenikinju, A. (2008). Efficiency of the Energy Sector and its Impact on the

Competitiveness of the Nigerian Economy. International Association

for Energy Economics, 27(32), 131-9.

Adjei-Mantey, K., & Takeuchi, K. (2023). Risk aversion and cleaner cooking

fuel choice: an empirical study in Ghana. Environment and

Development Economics, 28(2), 130-148.

117

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Ahmad, W., Prakash, R., Uddin, G. S., Chahal, R. J. K., Rahman, M. L., &

Dutta, A. (2020). On the intraday dynamics of oil price and exchange

rate: What can we learn from China and India? Energy Economics, 91,

104871.

Ahmad, I., Iqbal, S., Khan, S., Han, H., Vega-Muñoz, A., & Ariza-Montes, A.

(2022). Macroeconomic effects of crude oil shocks: Evidence from

South Asian countries. Frontiers in Psychology, 13.

Ahmed, F., Osama Daudpota, M., & Kashif, M. (2017). Oil price shocks and

industry level production using vector autoregression: Empirical

evidence from Pakistan. Global Journal of Business Research, 11(3),

13-25.

Aimer, N. M. (2016). The effects of fluctuations of oil price on economic

growth of Libya. Energy Economics Letters, 3(2), 17-29.

Akalpler, E., & BAKAR, A. N. (2018). The impact of oil price instability on

economic growth: Evidence from Nigeria. Business Economics and

Management Research Journal, 1(1), 39-53.

Alao, R. O., & Payaslioglu, C. (2021). Oil price uncertainty and industrial

production in oil-exporting countries. Resources Policy, 70, 101957.

Al-Risheq, S. M. (2016). The impact of oil prices on industrial production in

developing countries.

Al-Sasi, B. O., Taylan, O., & Demirbas, A. (2017). The impact of oil price

volatility on economic growth. Energy Sources, Part B: Economics,

Planning, and Policy, 12(10), 847-852.

118

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Altunbaş, Y., & Thornton, J. (2019). The impact of financial development on

income inequality: A quantile regression approach. Economics

Letters, 175, 51-56.

Amoako, G. K., Asafo-Adjei, E., Mintah Oware, K., & Adam, A. M. (2022).

Do volatilities matter in the interconnectedness between world energy

commodities and stock markets of BRICS? Discrete Dynamics in

Nature and Society, 2022.

Ang, J. S. (1992). On the theory of finance for privately held firms. The

Journal of Entrepreneurial Finance, 1(3), 185-203.

Appiah, G. K., Oduro, E., & Benn, S. (2020). Empirical analysis of crude oil

consumption and price on Ghana's economic growth. Open Science

Journal, 5(4).

Archer, C., Junior, P. O., Adam, A. M., Asafo-Adjei, E., & Baffoe, S. (2022).

Asymmetric Dependence between Exchange Rate and Commodity

Prices in Ghana. Annals of Financial Economics, 17(02), 2250012.

Armah, B. (2003). Economic analysis of the energy sector. Energy

Commission of Ghana publication. Accra, Ghana.

Armah, M., Amewu, G., & Bossman, A. (2022). Time-frequency analysis of

financial stress and global commodities prices: Insights from wavelet-

based approaches. Cogent Economics & Finance, 10(1), 2114161.

Asafo-Adjei, E., Adam, A. M., & Darkwa, P. (2021). Can crude oil price

returns drive stock returns of oil producing countries in Africa?

Evidence from bivariate and multiple wavelets. Macroeconomics and

Finance in Emerging Market Economies, 1-19.

119

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Asafo-Adjei, E., Adam, A. M., Arthur, C. L., Seidu, B. A., & Gyasi, R. M.

(2022). Similarities among equities returns in multi-frequencies:

insights from sustainable responsible investing. Journal of sustainable

finance & investment, 1-23.

Asafo-Adjei, E., Agyapong, D., Agyei, S. K., Frimpong, S., Djimatey, R., &

Adam, A. M. (2020). Economic policy uncertainty and stock returns of

Africa: a wavelet coherence analysis. Discrete Dynamics in Nature and

Society, 2020, 1-8.

Asafo-Adjei, E., Boateng, E., Isshaq, Z., Idun, A. A. A., Owusu Junior, P., &

Adam, A. M. (2021). Financial sector and economic growth amid

external uncertainty shocks: insights into emerging economies. Plos

one, 16(11), e0259303.

Awunyo-Vitor, D., Samanhyia, S., & Bonney, E. A. (2018). Do oil prices

influence economic growth in Ghana? An empirical analysis. Cogent

Economics & Finance, 6(1).

Aye, G. C., Dadam, V., Gupta, R., & Mamba, B. (2014). Oil price uncertainty

and manufacturing production. Energy Economics, 43, 41-47.

Balcilar, M., Gupta, R., & Pierdzioch, C. (2016). Does uncertainty move the

gold price? New evidence from a nonparametric causality-in-quantiles

test. Resources Policy, 49, 74-80.

Bans-Akutey, M., Yaw Deh, I., & Mohammed, F. (2016). What is the effect of

Inflation on Manufacturing Sector Productivity in Ghana?

120

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Barson, Z., Junior, P. O., Adam, A. M., & Asafo-Adjei, E. (2022).

Connectedness between Gold and Cryptocurrencies in COVID-19

Pandemic: A Frequency-Dependent Asymmetric and Causality

Analysis. Complexity, 2022.

Basher, S. A., & Sadorsky, P. (2006). Oil price risk and emerging stock

markets. Global finance journal, 17(2), 224-251.

Bassett Jr, G., & Koenker, R. (1978). Asymptotic theory of least absolute error

regression. Journal of the American Statistical Association, 73(363),

618-622.

Beji, S., & Belhadj, A. (2014). What are the determining factors of

Industrialization in Africa? Federalismi. it, 1-20.

Bildirici, M., Guler Bayazit, N., & Ucan, Y. (2020). Analyzing crude oil

prices under the impact of covid-19 by using lstargarchlstm. Energies,

13(11), 2980.

Bilgin, M. H., Gozgor, G., & Karabulut, G. (2015). The impact of world

energy price volatility on aggregate economic activity in developing

Asian economies. The Singapore Economic Review, 60(01), 1550009.

Bloomfield, D. S., McAteer, R. J., Lites, B. W., Judge, P. G., Mathioudakis,

M., & Keenan, F. P. (2004). Wavelet phase coherence analysis:

application to a quiet-sun magnetic element. The Astrophysical

Journal, 617(1), 623.

Boateng, E., Adam, A. M., & Junior, P. O. (2021). Modelling the

heterogeneous relationship between the crude oil implied volatility

index and African stocks in the coronavirus pandemic. Resources

policy, 74, 102389.

121

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Boateng, E., Asafo-Adjei, E., Addison, A., Quaicoe, S., Yusuf, M. A., &

Adam, A. M. (2022). Interconnectedness among commodities, the real

sector of Ghana and external shocks. Resources Policy, 75, 102511.

Bossman, A., Gherghina, Ş. C., Asafo-Adjei, E., Adam, A. M., & Agyei, S. K.

(2022). Exploring the asymmetric effects of economic policy

uncertainty and implied volatilities on energy futures returns: novel

insights from quantile-on-quantile regression. Journal of Business

Economics and Management, 23(6), 1351-1376.

Buabeng, E., Ayesu, E. K., & Adabor, O. (2019). The effect of exchange rate

fluctuation on the performance of manufacturing firms: Empirical

evidence from Ghana. Economics Literature, 1(2), 133-147.

Bugshan, A. (2021). The impact of oil price volatility on firm financial and

market performance: an empirical analysis from oil exporting

countries.

Burakov, D. (2016). Oil prices, exchange rate and prices for agricultural

commodities: Empirical evidence from Russia. AGRIS on-line Papers

in Economics and Informatics, 8(665-2016-45118), 33-47.

Butler, R. (2022). What Happens If Interest Rates Increase Too Quickly?

Retrieved: [Link]

[Link]

Cameron, R. (2009). A sequential mixed model research design: Design,

analytical and display issues. International journal of multiple

research approaches, 3(2), 140-152.

122

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Cantah, W., & Asmah, E. (2015). Crude oil price and growth of output: the

case of Ghana. International Journal of Economics, Commerce and

Management, 3(5).

Cantore, N., Antimiani, A., & Anciaes, P. R. (2012). Energy price shocks:

sweet and sour consequences in developing countries.

Centre for Study of Africa Economies (CSAE) (2014). Crude oil price and

economic growth: The case of Ghana. Conference paper.

Chang, L., Baloch, Z. A., Saydaliev, H. B., Hyder, M., & Dilanchiev, A.

(2022). Testing oil price volatility during Covid-19: Global economic

impact. Resources Policy, 78, 102891.

Chatziantoniou, I., Filippidis, M., Filis, G., & Gabauer, D. (2021). A closer

look into the global determinants of oil price volatility. Energy

Economics, 95, 105092.

Chen, H., Liu, L., Wang, Y., & Zhu, Y. (2016). Oil price shocks and US dollar

exchange rates. Energy, 112, 1036-1048.

Chen, S. S., & Hsu, K. W. (2012). Reverse globalization: Does high oil price

volatility discourage international trade? Energy Economics, 34(5),

1634-1643.

Chen, Y., Zhu, X., & Li, H. (2022). The asymmetric effects of oil price shocks

and uncertainty on non-ferrous metal market: Based on quantile

regression. Energy, 246, 123365.

Chow, G. C., & Lin, A. L. (1971). Best linear unbiased interpolation,

distribution, and extrapolation of time series by related series. The

review of Economics and Statistics, 372-375.

123

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Cologni, A., & Manera, M. (2008). Oil prices, inflation and interest rates in a

structural cointegrated VAR model for the G-7 countries. Energy

economics, 30(3), 856-888.

Cornell, B. (2018). What is the alternative hypothesis to market

efficiency?‖ Journal of Portfolio Management, 44(7), 3–6.

Coudert, V., Mignon, V., & Penot, A. (2007). Oil price and the dollar. Energy

Studies Review, 15(2).

Council on Foreign Relations (July, 2016). Oil Price Volatility: Causes,

Effects, and Policy Implications. Insights from a CFR Workshop.

Available: [Link]

effects-and-policy-implications.

Creswell, J. W. (2013). Steps in conducting a scholarly mixed methods study.

Creswell, J. W., & Plano Clark, V. L. (2011). Choosing a mixed methods

design. Designing and conducting mixed methods research, 2, 53-106.

Dadzie, P., Nambie, N. B., & Obobi, B. A. (2023). Impact of Petroleum

Energy Price Volatility on Commodity Prices in Ghana. International

Journal of Economics and Financial Issues, 13(1), 73.

Dawar, I., Dutta, A., Bouri, E., & Saeed, T. (2021). Crude oil prices and clean

energy stock indices: Lagged and asymmetric effects with quantile

regression. Renewable Energy, 163, 288-299.

Demir, A., Pesqué-Cela, V., Altunbas, Y., & Murinde, V. (2022). Fintech,

financial inclusion and income inequality: a quantile regression

approach. The European Journal of Finance, 28(1), 86-107.

Dogah, K. E. (2015). The impact of oil price shocks on the macroeconomy of

Ghana. Journal of Poverty, Investment and Development, 9, 37-54.

124

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Donkor, R. A. (2018). The Impact of Oil Price Volatility on Exchange Rate

Volatility: A Study of Some Oil Dependent Economies (Doctoral

dissertation, University of Ghana).

Dramani, J. B., & Frimpong, P. B. (2020). The effect of crude oil price shocks

on macroeconomic stability in Ghana. OPEC Energy Review, 44(3),

249-277.

Eagle, B. (2017). Oil price volatility and macroeconomy: Tales from top two

oil producing economies in Africa. Journal of Economic & Financial

Studies, 5(04), 45-55.

Eksi, I. H., Izgi, B. B., & Senturk, M. (2011). Reconsidering the relationship

between oil prices and industrial production: Testing for cointegration

in some of the OECD countries. Eurasian Journal of Business and

Economics, 4(8), 1-12.

Economist Intelligence. (June, 2015). Ghana to remove remaining fuel

subsidies.[Link]

Country=Ghana&topic=Economy&subtopic=For 1.

Energy Commission of Ghana. (2021). National energy statistics 2020. Accra.

Energy Information Administration. (2021). Use of energy explained. Energy

consumption in industry. Retrieved from [Link]

energyexplained/use-of-energy/[Link]

Energy Information Administration. (2022). "What Drives Crude Oil Prices?"

Retrieved from [Link]

php

Fauzel, S., Seetanah, B., and Sannasee, R. V. (2015). Productivity Spillovers

of FDI in the Manufacturing Sector of Mauritius. Evidence from a

125

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Dynamic Framework. The Journal of Developing Areas, Vol. 49, No.2,

Spring 2015.

Feng, Y., Wang, X., Chen, S., & Liu, Y. (2022). Impact of Oil

Financialization on Oil Price

Ftiti, Z., Guesmi, K., Teulon, F., & Chouachi, S. (2016). Relationship between

crude oil prices and economic growth in selected OPEC

countries. Journal of Applied Business Research (JABR), 32(1), 11-22.

Fluctuation: A Perspective of Heterogeneity. Energies, 15(12), 4294.

Fosu, A. K., & Aryeetey, E. (2008). Ghana's post-independence economic

growth 1960-2000. The Economy of Ghana: Analytical Perspective on

Stability, Growth & Poverty, 36-77.

Ghana Statistical Service. (2021). Rebased 2013-2019 Annual Gross Domestic

Product. Accra

Ghana Statistical Service. (2020). Rebased 2013-2019 Annual Gross Domestic

Product. Accra

Gong, X. L., Feng, Y. K., Liu, J. M., & Xiong, X. (2023). Study on

international energy market and geopolitical risk contagion based on

complex network. Resources Policy, 82, 103495.

Gong, X.; Lin, B.Q. Time-varying effects of oil supply and demand shocks on

China’s macro-economy. Energy 2018, 149, 424– 437.

González, A., & Nabiyev, S. (2009). Oil price fluctuations and its effect on

GDP growth.

Graham, E., Ackah, I., & Gyampo, R. E. (2016). Politics of oil and gas in

Ghana. Insight on Africa, 8(2), 131-141.

126

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Gujarati, D. N., & Porter, D. C. (2009). Multicollinearity: What Happens If the

Regressors Are Correlated? In Basic econometrics (Fifth Edit, pp.

320–364). McGraw-Hill.

Gyampo, R. (2010). Saving Ghana from its oil: a critical assessment of

preparations so far made. African Research Review, 4(3).

Hamdi, B., Aloui, M., Alqahtani, F., & Tiwari, A. (2019). Relationship

between the oil price volatility and sectoral stock markets in oil-

exporting economies: Evidence from wavelet nonlinear denoised based

quantile and Granger-causality analysis. Energy Economics, 80, 536-

552.

Hamilton, J. D. (1983). Oil and the macroeconomy since World War

II. Journal of political economy, 91(2), 228-248.

Hamilton, J. D. (2009). Causes and Consequences of the Oil Shock of 2007-

08. National Bureau of Economic Research.

Hau, L., Zhu, H., Huang, R., & Ma, X. (2020). Heterogeneous dependence

between crude oil price volatility and China’s agriculture commodity

futures: Evidence from quantile-on-quantile regression. Energy, 213,

118781.

Hennink, M., Hutter, I., & Bailey, A. (2020). Qualitative research methods.

Sage.

Henriques, I., & Sadorsky, P. (2011). The effect of oil price volatility on

strategic investment. Energy Economics, 33(1), 79-87.

Herrera, A. M., Lagalo, L. G., & Wada, T. (2011). Oil price shocks and

industrial production: Is the relationship linear? Macroeconomic

dynamics, 15(S3), 472-497.

127

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Holden, M. T., & Lynch, P. (2004). Choosing the appropriate methodology:

Understanding research philosophy. The marketing review, 4(4), 397-

409.

Hong, M., Wang, X., & Li, Z. (2022). Will Oil Price Volatility Cause Market

Panic? Energies, 15(13), 4629.

Hooker, M. A. (1996). This is what happened to the oil price-macroeconomy

relationship: Reply. Journal of Monetary Economics, 38(2), 221-222.

Hooker, M. A. (2002). Are oil shocks inflationary? Asymmetric and nonlinear

specifications versus changes in regime. Journal of Money, Credit and

Banking, 34(2), 540-561.

Humaira, Y., Ying, W., Hashim, Z., & Yasir, A. S. (2019). Does oil price

volatility influence real sector growth? Empirical evidence from

Pakistan. Energy Reports, 5, 688-703.

Humbatova, S. I. (2019). The impact of oil factor on Azerbaijan economy.

Humssi, Petrovskaya, & Abueva, (2022). Modelling the Impact of World Oil

Prices and the Mining and Quarrying Sector on the United Arab

Emirates’ GDP. Mathematics, 11(1), 94.

Iddrisu, A. A., Adam, B., & Halidu, B. O. (2015). The influence of foreign

direct investment (FDI) on the productivity of the industrial sector in

Ghana. International Journal of Academic Research in Accounting,

Finance and Management Sciences, 5(3), 1-13

Idun, A. A. A., Asafo-Adjei, E., Adam, A. M., & Isshaq, Z. (2022). Dynamic

Connectedness between Indicators of the Ghana Stock Exchange

Returns and Macroeconomic Fundamentals. Risks, 10(11), 215.

128

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Iganiga, B. O., Anyanwu, U. N., Ikubor, J. O., & Ojima, D. (2021). Oil Price

Shocks and Industrial Output in Nigeria; is The Relationship Linear?

Ijirshar, V. U. (2015). The empirical analysis of agricultural exports and

economic growth in Nigeria. Journal of Development and Agricultural

Economics, 7(3), 113-122.

International Energy Agency. (2011). Key World Energy Statistics. Tech. rep.,

International Energy Agency. URL [Link] org/textbase/nppdf/

free/2011/key_world_energy_stats. pdf.

International Monetary Fund (2020) Balance of Payments database,

supplemented by data from the United Nations Conference on Trade

and Development and official national sources.

Jawadi, F., & Ftiti, Z. (2019). Oil price collapse and challenges to economic

transformation of Saudi Arabia: A time-series analysis. Energy

Economics, 80, 12-19.

Jena, S. K., Tiwari, A. K., Hammoudeh, S., & Roubaud, D. (2019).

Distributional predictability between commodity spot and futures:

Evidence from nonparametric causality-in-quantiles tests. Energy

Economics, 78, 615-628.

Jeong, K., Härdle, W. K., & Song, S. (2012). A consistent nonparametric test

for causality in quantile. Econometric Theory, 28(4), 861-887.

Jiménez-Rodrguez , R., & Sánchez, M. (2005). Oil price shocks and real GDP

growth; empirical evidence for some. Applied Economics, 37, 201-228.

Jiranyakul, K. (2016). The response of industrial production to the price of oil:

new evidence from Thailand. Munich Personal RePEc Archive.

129

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Joo, Y. C., & Park, S. Y. (2017). Oil prices and stock markets: does the effect

of uncertainty change over time? Energy Economics, 61, 42-51.

Kasongwa, L., & Minja, E. (2022). Moderating effects of oil price on the

impact of macroeconomic variables on stock market

performance. Business management review, 25(1), 75-89.

Kassem, M., Khoiry, M. A., & Hamzah, N. (2019). Risk factors in oil and gas

construction projects in developing countries: A case

study. International Journal of Energy Sector Management, 13(4),

846-861.

Kassim, F., & Isik, A. (2020). Impact of Energy Consumption on Industrial

Growth in a Transition Economy: Evidence from Nigeria.

Kilian, L., & Murphy, D. P. (2014). The role of inventories and speculative

trading in the global market for crude oil. Journal of Applied

econometrics, 29(3), 454-478.

Kjellen, J., & Lund, A. (2022). From Tartous to Tobruk: The Return of

Russian Sea Power in the Eastern Mediterranean. FOI-Swedish

Défense Research Agency.

Koenker, R., & Bassett Jr, G. (1978). Regression quantiles. Econometrica:

journal of the Econometric Society, 33-50.

Koirala, N. P., & Ma, X. (2020). Oil price uncertainty and US employment

growth. Energy Economics, 91.

Kojima, M. (2016). Fossil fuel subsidy and pricing policies: recent developing

country experience. World Bank Policy Research Working Paper,

(7531).

130

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Kriskkumar, K., & Naseem, N. A. M. (2019). Analysis of oil price effect on

economic growth of ASEAN net oil exporters. Energies, 12(17), 3343

Le, T. H., & Chang, Y. (2015). Effects of oil price shocks on the stock market

peformance: Do nature of shocks and economies matter? Energy

Economics, 51, 261-274.

Lewis, W.A. (1954). Economic Development with Unlimited Supplies of

Labour. Manchester School, 28, 139-91.

Li, Z.; Huang, Z.; Failler, P. (2022). Dynamic Correlation between Crude Oil

Price and Investor Sentiment in China: Heterogeneous and

Asymmetric Effect. Energies 2022, 15, 687.

Li, R., Li, S., Wang, Y., & Li, G. (2016). How does oil price volatility affect

non-energy sector stock prices in China? Evidence from a quantile

regression approach. Energy Economics, 54, 283-291.

Lin, B., & Wesseh Jr, P. K. (2013). What causes price volatility and regime

shifts in the natural gas market. Energy, 55, 553-563.

Lin, B., Wesseh, P. K., & Appiah, M. O. (2014). Oil price fluctuations,

volatility spillover, and the Ghanaian equity market: Implication for

portfolio management and hedging effectiveness. Energy Economics,

172-182.

Lo, A. W. (2004). The adaptive markets hypothesis. The Journal of Portfolio

Management, 30(5), 15-29.

Lux, T., Segnon, M., & Gupta, R. (2015). Modeling and forecasting crude oil

price volatility: Evidence from historical and recent data (No. 31).

FinMaP-Working Paper.

131

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Manteaw, S. (2011). Making oil and gas governance work for democratic

development—A civil society perspective from Ghana. Fuelling the

world—Failing the region, 182-91.

Mgbame, C. O., & Donwa, P. A. (2015). The Impact of Oil Price Volatility on

Economic Growth: Conceptual perspective. Journal of

Multidisciplinary Research and Development, 2(9), 80-85.

Mo, B., Chen, C., Nie, H., & Jiang, Y. (2019). Visiting effects of crude oil

price on economic growth in BRICS countries: fresh evidence from

wavelet-based quantile-on-quantile tests. Energy, 178, 234-251.

Moussa, B., Amadu, I., Idrissa, O., & Abdou, B. (2019). The impact of foreign

direct investment on the productivity of manufacturing firms in

Cameroon. Journal of Economics and Development Studies, 7(1), 25-

34.

Müller, U. A., Dacorogna, M. M., Davé, R. D., Olsen, R. B., Pictet, O. V., &

Von Weizsäcker, J. E. (1997). Volatilities of different time

resolutions—analyzing the dynamics of market components. Journal

of Empirical Finance, 4(2-3), 213-239.

Mukhtarov, S., Humbatova, S., Mammadli, M., & Hajiyev, N. G. O. (2021).

The impact of oil price shocks on national income: Evidence from

Azerbaijan. Energies, 14(6), 1695.

Nchor, D., Klepáč, V., & Adamec, V. (2016). Effects of oil price shocks on

the Ghanaian economy. Acta Universitatis Agriculturae et Silviculturae

Mendelianae Brunensis, 64(1), 315-324.

Nkrumah-Boadu, B., Owusu Junior, P., Adam, A., & Asafo-Adjei, E. (2022).

Safe haven, hedge and diversification for African stocks:

132

Digitized by Sam Jonah Library

University of Cape Coast [Link]

cryptocurrencies versus gold in time-frequency perspective. Cogent

Economics & Finance, 10(1), 2114171.

Obite, C. P., Olewuezi, N. P., Ugwuanyim, G. U., & Bartholomew, D. C.

(2020). Multicollinearity effect in regression analysis: A feed forward

artificial neural network approach. Asian journal of probability and

statistics, 6(1), 22-33.

Ogiri, I. H., Amadi, S. N., Uddin, M. M., & Dubon, P. (2013). Oil price and

stock market performance in Nigeria: An empirical analysis. American

Journal of Social and Management Sciences, 4(1), 20-41.

Ogunsakin, S., & Oloruntuyi, A. O. (2017). Oil price volatility and

macroeconomic performance in two top net oil producing countries in

Africa. International Economic Research, 8(3), 59-79.

Okwu, AT, Akpa, EO, Oseni, IO and Obiakor, RT (2020). Oil Export Revenue

and Exchange Rate: An Investigation of Asymmetric Effects on

Households’ Consumption Expenditure in Nigeria, Babcock Journal of

Economics, 7, 13-28.

Ormston, R., Spencer, L., Barnard, M., & Snape, D. (2014). The foundations

of qualitative research. Qualitative research practice: A guide for

social science students and researchers, 2(7), 52-55.

O'Neill, A. (December, 2022). Ghana: Share of economic sectors in the gross

domestic product (GDP) from 2011 to 2021. [Link]

com/statistics/447524/share-of-economic-sectors-in-the-gdp-in-

ghana/#:~:text=This%20statistic%20shows%20the%20

share,sector%20contributed%20about%2045.93%20percent.

133

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Osei, D. (2011, June 17). A concise history of oil and gas exploration in

Ghana. Retrieved April 21, 2020, from

[Link] .info/Article/A-CONCISE-HISTORY-OF-OIL-

AND-GAS-EXPLORATION-IN-GHANA/552309.

Osei, P. M., & Adam, A. M. (2020). Quantifying the information flow

between Ghana stock market index and its constituents using transfer

entropy. Mathematical Problems in Engineering, 2020, 1-10.

Owusu Junior, P., Frimpong, S., Adam, A. M., Agyei, S. K., Gyamfi, E. N.,

Agyapong, D., & Tweneboah, G. (2021). COVID-19 as information

transmitter to global equity markets: evidence from CEEMDAN-based

transfer entropy approach. Mathematical Problems in Engineering,

2021, 1-19.

Oteng-Abayie, E. F., Dramani, J. B., Sulemana, M., & Adusah-Poku, F.

(2023). The Asymmetric Effects of Oil Price Shocks on Aggregate

Demand for Goods and Services in Ghana. Journal of Energy, 2023.

Phan, D. H. B., Tran, V. T., & Nguyen, D. T. (2019). Crude oil price

uncertainty and corporate investment: New global evidence. Energy

Economics, 77, 54-65.

Phan, D. H. B., Tran, V. T., Nguyen, D. T., & Le, A. (2020). The importance

of managerial ability on crude oil price uncertainty-firm performance

relationship. Energy Economics, 88, 104778.

Prabheesh, K. P., Padhan, R., & Garg, B. (2020). COVID-19 and the oil price–

stock market nexus: Evidence from net oil-importing countries. Energy

Research Letters, 1(2), 13745.

134

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Riaz, M. F., Sial, M. H., & Nasreen, S. (2016). Impact of oil price volatility on

manufacturing production of Pakistan. Bulletin of Energy Economics, 4

(1), 23-34.

Ramberg, D. J., Chen, Y. H., Paltsev, S., & Parsons, J. E. (2017). The

economic viability of gas-to-liquids technology and the crude oil–

natural gas price relationship. Energy Economics, 63, 13-21.

Ramelli, S., & Wagner, A. F. (2020). Feverish stock price reactions to

COVID-19. The Review of Corporate Finance Studies, 9(3), 622-655.

Salim, R., & Rafiq, S. (2011). The impact of crude oil price volatility on

selected Asian emerging economies. In Proceedings of global business

and social science research conference (pp. 1-33). World Business

Institute Australia.

Saqib, N., Masnoon, M., & Rafique, N. (2013). Impact of foreign direct

investment on economic growth of Pakistan. Advances in Management

& Applied Economics, 3(1), 35-45.

Saunders, M., Lewis, P. & Thornhill, A. (2012). Research Methods for

Business Students. 6th edition, Pearson Education Limited.

Shaari, M. S., Pei, T. L., & Rahim, H. A. (2013). Effects of oil price shocks on

the economic sectors in Malaysia. International Journal of Energy

Economics and Policy, 3(4), 360-366.

Simon, G., Lee, J. A., Cottrell, M., & Verleysen, M. (2007). Forecasting the

CATS benchmark with the Double Vector Quantization

method. Neurocomputing, 70(13-15), 2400-2409.

Singh, S., Bansal, P., & Bhardwaj, N. (2022). Correlation between geo

political risk, economic policy uncertainty and exchange rates using

135

Digitized by Sam Jonah Library

University of Cape Coast [Link]

partial and multiple wavelet coherence in P5 nations. International

Journal of Global Warming, 27(4), 379-401.

Slevitch, L. (2011). Qualitative and quantitative methodologies compared:

Ontological and epistemological perspectives. Journal of quality

assurance in hospitality & tourism, 12(1), 73-81.

Smith, J. K. (1983). Quantitative versus qualitative research: An attempt to

clarify the issue. Educational researcher, 12(3), 6-13.

Sorrell, S. (2015). Reducing energy demand: A review of issues, challenges

and approaches. Renewable and Sustainable Energy Reviews, 47, 74-

82.

Taghizadeh-Hesary, F., Rasolinezhad, E., & Kobayashi, Y. (2015). Oil price

flutuations and oil consuming sectors: An empirical analysis of Japan.

ADBI Working Paper.

Takyiwa, J. A. (2014). Oil discovery, In-migration and infrastructural

challenges in the Sekondi-Takoradi metropolis (Doctoral dissertation,

University Of Ghana).

Tang, W., Wu, L., & Zhang, Z. (2010). Oil price shocks and their short-and

long-term effects on the Chinese economy. Energy Economics, 32, S3-

S14.

Taofik, I. (2018). Oil price fluctuation and aggregate output performance in

Nigeria. Munich Peraonal RePEc Archive(88636), 2-10.

Tapsın, G. (2017). The link between industry value added and electricity

consumption. European Scientific Journal, 13(13), 41-56.

136

Digitized by Sam Jonah Library

University of Cape Coast [Link]

The Business and Financial Times. (October, 2022). Association of Ghana

Industries (AGI) laments tough business environment. https://

[Link]/2022/10/13/agi-laments-tough-business-environment/

Tissaoui, K., Zaghdoudi, T., Hakimi, A., & Nsaibi, M. (2022). Do Gas Price

and Uncertainty Indices Forecast Crude Oil Prices? Fresh Evidence

Through XGBoost Modeling. Computational Economics, 1-25.

Torrence, C., & Compo, G. P. (1998). A Practical Guide to Wavelet Analysis.

Bulletin of the American Meteorological Society, 61-78. doi:10.1175/

1520-0477(1998)079<0061:APGTWA>[Link];2

Torul, O., & Alper, C. E. (2010). Asymmetric effects of oil prices on the

manufacturing sector in Turkey. Review of Middle East Economics and

Finance, 6(1), 90-105.

Tsatsu, I. K. (2022). The Impact of Escalating Fuel Prices on Businesses in

Ghana. The Korea Institute for International Economic Policy.

[Link]

brdctsFileNo=82673.

Tunyo, D. A., Armah, M., Cantah, W. G., & Suleman, S. (2021). Crude oil

production and macroeconomic performance in Ghana. OPEC Energy

Review, 45(4), 475-505.

Tverberg, G. E. (2012). Oil supply limits and the continuing financial

crisis. Energy, 37(1), 27-34.

Walfure, O. G., & Nurudeen, A. (2010). Determinants of FDI in Nigeria, an

empirical analysis. Global Journal of Human Social Sciences, 10(1),

26-34.

137

Digitized by Sam Jonah Library

University of Cape Coast [Link]

Wang, J., Niu, X., Zhang, L., Liu, Z., & Wei, D. (2022). The influence of

international oil prices on the exchange rates of oil exporting countries:

Based on the hybrid copula function. Resources Policy, 77, 102734.

Wen, F., Min, F., Zhang, Y. J., & Yang, C. (2019). Crude oil price shocks,

monetary policy, and China's economy. International Journal of

Finance & Economics, 24(2), 812-827.

Yazdani, M., & Noor, A. (2015). Evaluation of the effect of the fluctuations in

oil prices and production gap on the trade balance of Iran. Journal of

Economic Science, 9(31.), 31-40.

Yilmazkuday, H. (2021). COVID-19 effects on the S&P 500 index. Applied

Economics Letters, 1-7.

Yu, Y., Guo, S., & Chang, X. (2022). Oil prices volatility and economic

performance during COVID-19 and financial crises of 2007–

2008. Resources policy, 75, 102531.

Zhang, C., Mou, X., & Ye, S. (2022). How do dynamic jumps in global crude

oil prices impact China's industrial sector?.Energy, 249, 123605.

Zhang, Z., Wang, Y., Xiao, J., & Zhang, Y. (2023). Not all geopolitical shocks

are alike: Identifying price dynamics in the crude oil market under

tensions. Resources Policy, 80, 103238.

Zhu, H., Guo, Y., You, W., & Xu, Y. (2016). The heterogeneity dependence

between crude oil price changes and industry stock market returns in

China: Evidence from a quantile regression approach. Energy

Economics, 55, 30-41.

138

Digitized by Sam Jonah Library

University of Cape Coast [Link]

APPENDICES
APPENDIX A
The realised crude oil price volatility was extracted in logarithmic

returns through the Generalized Autoregressive Conditional

Heteroskedasticity (GARCH (1,1)) model. The GARCH model is a suitable

choice for computing crude oil price volatility in this study because it

explicitly accounts for the time-varying and non-constant nature of volatility

in financial and commodity markets, allowing us to capture the dynamic and

evolving characteristics of crude oil price fluctuations, which are essential for

a comprehensive empirical analysis of their impact on industrial output in

Ghana. The GARCH (1,1) model is a time series model commonly used in

finance to estimate the volatility of financial assets. It is an extension of the

Autoregressive Conditional Heteroskedasticity (ARCH) model, which allows

for the conditional variance to be dependent on lagged values of the squared

residuals. The development of the Generalized ARCH (GARCH) model aimed

to address the difficulties faced by ARCH models.

Similar to the ARCH model, the GARCH model employs a weighted

mean of past squared residuals, but its weights gradually decrease rather than

becoming zero. According to the model, the most effective method for

predicting the next period's variance involves a weighted combination of the

long-term average variance, the predicted variance for the current period, and

the most recent squared residual, which represents new information for the

139

Digitized by Sam Jonah Library

University of Cape Coast [Link]

current period (Lux, Segnon, & Gupta, 2015). Existence of volatility from the

data is also ascertained.

(16)

where:

is the conditional variance of the return’s series at time .

is a constant representing the long-run average of the conditional variance.

is the logarithmic returns,

is the coefficient measuring the impact of the squared returns of the previous

period ( ) on the current period’s conditional variance .

β is the coefficient measuring the impact of the past period's conditional

variance ( ) on the current period's conditional variance .

To calculate the monthly realised volatility , by taking the square root of

the sum of squared returns over a month:

√ ∑ (17)

where ranges from 1 to n, and n is the number of returns in the period, and

is the average return of the return’s series

The estimated conditional variance, , to the realised volatility, , to assess

the accuracy of the GARCH (1,1) model in capturing the volatility of the

return’s series.

The study shows the GARCH (1,1) model to estimate the volatility of

crude oil price. This can be presented as

(18)

where:

is the logarithmic returns of crude oil price

, , and β is as defined previously.

140

Digitized by Sam Jonah Library

University of Cape Coast [Link]

To calculate the monthly realised volatility , by taking the square

root of the sum of squared returns over a month:

√ ∑ (13)

where ranges from 1 to n, and n is the number of returns in the period, and

is the average return of the returns series OP

141

Digitized by Sam Jonah Library

 University of Cape Coast [Link]

APPENDIX B
Table 10: Correlation Matrix
CONS ELECT ISO MFG MQ WS ROVX EC FDI INT OP
CONS 1.000
ELECT 0.975** 1.000
ISO 0.992** 0.986** 1.000
MFG 0.985** 0.985** 0.996** 1.000
MQ 0.983** 0.975** 0.994** 0.983** 1.000
WS 0.973** 0.999** 0.986** 0.987** 0.974** 1.000
ROVX 0.119 0.131* 0.124 0.138* 0.112 0.136* 1.000
EC 0.913** 0.902** 0.921** 0.929** 0.909** 0.910** 0.186** 1.000
FDI 0.830** 0.751** 0.795** 0.809** 0.754** 0.751** 0.011 0.769** 1.000
INT -0.638** -0.608** -0.667** -0.705** -0.641** -0.625** -0.101 -0.661** -0.716** 1.000
OP 0.451** 0.313** 0.414** 0.417** 0.399** 0.314** -0.259** 0.403** 0.760** -0.612** 1.000
Source: Author’s Construct (2023)

Note: The variables CONS, EC, ELECT, FDI, INT, ISO, MFG, MQ, OP, ROVX and WS represent Construction, Energy consumption,
Electricity, Foreign Direct Investment, interest rate, Industrial sector output, Manufacturing, Mining and Quarrying, Crude oil price, Realised
crude oil volatility, and Water and Sewerage respectively. Also, **, and * denotes significance at 1% and 5% respectively.

142

Digitized by Sam Jonah Library