G R E A T Z I M B A B W E U N I V E R S I T Y

MUNHUMUTAPA SCHOOL OF COMMERCE

DEPARTMENT OF ACCOUNTING AND INFORMATION SYSTEMS

“MODELLING THE EFFECTS OF TAXATION ON ECONOMIC GROWTH IN

ZIMBABWE FROM 2000 T0 2021”

By

Lasten Mudzingwa (M168376)

SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS OF

THE OF MASTER OF COMMERCE IN PROFESSIONAL ACCOUNTING
AND CORPORATE GOVERNANCE DEGREE

JULY 2022

MASVINGO, ZIMBABWE.

__________________________________________________________________

i
“Modelling the Effects of Taxation on Economic Growth in Zimbabwe from 2000 to 2021”

By

Lasten Mudzingwa

Dissertation submitted in partial fulfilment of the requirements for the degree of Master of
Commerce in Professional Accounting and Corporate Governance

Department of Accounting and Information Systems

Great Zimbabwe University

Supervisor: F. Tagwireyi

2022

ii
 DECLARATION AND COPYRIGHTS

I, Lasten Mudzingwa, do hereby declare that this dissertation is the result of my investigation
and research and that this has not been submitted in part or in full for any other University.

………………………………………. …………………………………………………..

L. Mudzingwa Date

Copyright © and Moral Rights for this dissertation are retained by the author and the Great
Zimbabwe University. This dissertation cannot be reproduced or quoted extensively from
without first obtaining permission in writing from the copyright holder(s). The content must not
be changed in any way or sold commercially in any format or medium without the formal
permission of the copyright holder(s).

iii
 APPROVAL FORM FOR POST GRADUATE PROGRAMME

The undersigned certify that they have supervised the student Lasten Mudzingwa's dissertation
entitled “Modelling the Effects of Taxation on Economic Growth in Zimbabwe from 2000
to 2021”: submitted in Partial fulfilment of the requirements of the Master of Commerce Degree
in Professional Accounting and Corporate Governance at Great Zimbabwe University.

…………………………………… ……………………………

Student Date

…………………………………… ……………………………

Supervisor Date

…………………………………… ……………………………

Chairperson Date

iv
 DEDICATIONS

I dedicate this research project to my late father, Gibson. C. Mudzingwa. You departed the time I
needed you the most, but you will always remain my pillar of strength.

v
 ACKNOWLEDGEMENTS

My foremost gratitude goes to the Almighty God for his permissive will which has enabled me
to successfully complete this piece of work in the face of a plethora of upheavals that could have
deterred me. I also extend my unequivocal gratitude to my supervisor Mr F. Tagwireyi and all
the lecturers in the Faculty of Accounting and Information Systems at Great Zimbabwe
University. Their unconditional commitment and unflinching dedication towards upgrading
knowledge and skills of students who are employees from different organisations is greatly
appreciated.

vi
 ABSTRACT

The study set out to investigate the effect of taxation on economic growth in Zimbabwe for the
time period 2000 to 2021 using time series data. This was prompted by the low levels of
economic growth prevailing in the country despite ZIMRA reporting on several occasions that it
is surpassing its tax revenue targets. The study adopted the Classical Linear Regression Model in
data analysis. Income tax, VAT, capital gains tax, inflation, interest rate and FDI were used as
independent variables while the dependent variable economic growth was proxied by GDP. The
study performed diagnostic tests such as multicollinearity, heteroscedasticity, autocorrelation,
unit root and stability tests with the main view of avoiding spurious results. Regression tests run
revealed that approximately 93.53 percent of the growth of the Zimbabwean economy is
influenced by income tax, VAT and interest rate. Regression results showed that income tax and
VAT have a positive significant effect on economic growth while interest rate has a negative
significant effect on economic growth. As such; the study recommended authorities to widen tax
base instead of increasing taxation in order to boost economic growth.

vii
 TABLE OF CONTENTS
Content Page Number
Title Page i
Declaration and Copyrights ii
Approval Form iii
Dedications iv
Acknowledgements v
Abstract vi
Table of Contents vii
List of Acronyms viii
List of Tables xiv
List of Appendices xv
List of Figures xvi

CHAPTER I
INTRODUCTION

1.0 INTRODUCTION 01
1.1 BACKGROUND TO THE STUDY 01
1.2 STATEMENT OF THE PROBLEM 04
1.3 JUSTIFCATION OF THE STUDY 05
1.4.1 GENERAL OBJECTIVES 05
1.4.2 SPECIFIC OBJECTIVES 05
1.5 RESEARCH QUESTIONS 06
1.6 RESEARCH HYPOTHESES 06
1.7 SIGNIFICANCE OF THE STUDY 07
1.8 LIMITATIONS OF THE STUDY 08
1.9 ASSUMPTIONS OF THE STUDY 08
1.10 DELIMITATIONS OF THE STUDY 08
1.11 DEFINITATION OF THE STUDY 09

viii
1.12 ACRONYMS AND ABBREVIATIONS 09
1.13 ORGANISATION OF THE STUDY 10
1.14 SUMMARY 10

CHAPTER II
LITERATURE REVIEW

1.0 INTRODUCTION 11
2.1CONCEPTUAL FRAMEWORK 11
2.1.1 TAXATION 11
2.1.2.1 INCOME TAX 13
2.1.2.2 VALUE ADDED TAX 13
2.1.2.3 CAPITAL GAINS TAX 14
2.1.3 TAX REVENUE 14
2.1.4 ECONOMIC GROWTH 15
2.1.5 INFLATION 16
2.1.6 INTEREST RATE 17
2.1.7 FOREIGN DIRECT INVESTMENT 19
2.1.7 TAXATION AND ECONOMIC GROWTH 20
2.2 THEORETICAL LITERATURE REVIEW 22
2.2.1 BENEFIT RECEIVED THEORY 22
2.2.2 OPTIMAL TAX THEORY 23
2.2.3 THE EXPEDIENCY THEORY OF TAXATION 25
2.2.4 ABILITY TO PAY THEORY 26
2.3 EMPIRICAL LITERATURE REVIEW 27
2.4 RESEARCH GAP 33
2.5 CHAPTER SUMMARY 33

CHAPTER III
RESEARCH METHODOLOGY

ix
3.0 INTRODUCTION 34
3.1RESEARCH PARADIGM 34
3.2 RESEARCH DESIGN 35
3.3 TARGET POPULATION AND SAMPLE DESIGN 36
3.4 DATA COLLECTION METHODS USED 36
3.5 THEORECTICAL FRAMEWORK 37
3.6.1 EMPIRICAL MODEL TO BE ESTIMATED 37
3.6.2 MATHEMATICAL AND STATISTICAL MODEL SPECIFICATION 37
3.7.1 MODEL JUSTIFICATION 39
3.7.2 FORM OF REGRESSION MODEL AND ANALYSIS 39
3.7.3 JUSTIFICATION OF VARIANCE 41
3.7.3.1 DEPENDENT VARIABLE 41
3.7.3.2 VALUE ADDED TAX 41
3.7.3.3 INCOME TAX 42
3.7.3.4 CAPITAL GAINS TAX 42
3.7.3.5 INFLATION 43
3.7.3.6 FOREIGN DIRECT INVESTMENT 43
3.7.3.7 INTEREST RATE 44
3.8 METHODS OF DATA ANALYSIS USED 45
3.8.1 DESCRIPTIVE STATISTICS 45
3.8.2 HYPOTHESES TESTING 46
3.8.3 DIAGNOSTIC TESTS 46
3.8.3.1 UNIT ROOT TEST 46
3.8.3.2 HETEROSCEDASTICITY TEST 47
3.8.3.3 MULTICOLLINEARITY TEST 48
3.8.3.4 AUTOCORRELATION TEST 48
3.8.3.5 STABILITY DIAGNOSTICS TEST 49
3.9 REGRESSION ANALYSIS 49
3.10 RESEARCH INSTRUMENTS 50
3.11 DATA PRESENTATION AND ANALYSIS PLAN 50

x
3.12 SUMMARY 50
CHAPTER IV
DATA PRESENTATION, ANALYSIS AND DISCUSSION

4.0 INTRODUCTION 52
4.1DESCRIPTIVE STATISTICS FOR ECONOMIC GROWTH AND ITS 52
INFLUENCING FACTORS
4.2 HYPOTHESES TEST RESULTS FOR ECONOMIC GROWTH AND ITS 53
INFLUENCING FACTORS
4.3.0 DIAGNOSTICS TEST RESULTS FOR FACTORS THAT INFLUENCE 55
THE GROWTH OF THE ZIMBABWEAN ECONOMY
4.3.1 UNIT ROOT TEST FOR FACTORS INFLUENCING ECONOMIC 55
GROWTH IN ZIMBABWE
4.3.2 HETEROSCEDASTICITY TEST RESULTS FOR ECONOMIC GROWTH 56
AND INFLUENCING FACTORS
4.3.3 MULTICOLLINEARITY TEST RESULTS FOR ECONOMIC GROWTH 57
AND INFLUENCING FACTORS
4.3.4 AUTOCORRELATION TEST RESULTS ON ECONOMIC GROWTH AND 59
THE FACTORS THAT AFFECT ITS GROWTH
4.3.5 STABILITY DIAGNOSTICS TEST RESULTS FOR ECONOMIC 60
GROWTH AND FACTORS RESPONSIBLE FOR ITS GROWTH IN
ZIMBABWE
4.4.0 ANALYSIS OF THE EFFECT OF TAXATION AND OTHER 61
MACROECONOMIC FUNDAMENTALS ON ECONOMIC GROWTH
REGRESSION RESULTS
4.4.1 INTERPRETATION OF R-SQUARED 64
4.4.2 INTERPRETATION OF ADJUSTED R-SQUARED 64
4.4.3 INTERPRETATION OF THE F-STATISTIC 64
4.4.4 INTERPRETATION OF REGRESSION RESULTS 64
4.4.5 INCOME TAX AND GROWTH OF THE ZIMBABWEAN ECONOMY 65

xi
4.4.6 THE RELATIONSHIP BETWEEN VAT AND ECONOMIC GROWTH 66
4.4.7 THE RELATIONSHIP BETWEEN INTEREST RATE AND ECONOMIC 67
GROWTH
4.5 SUMMARY 68

CHAPTER V
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

5.0 INTRODUCTION 69
5.1 SUMMARY OF THE STUDY 69
5.2 SUMMARY OF KEY FINDINGS ON THE EFFECT OF TAXATION ON 71
ECONOMIC GROWTH
5.3 HYPOTHESES TEST FINDINGS 73
5.4 CONCLUSIONS 74
5.5 RECOMMENDATIONS 75
5.6 SUGGESTIONS FOR FURTHER RESEARCH 75
REFERENCES 77
APPENDICES 86

xii
 LIST OF ACRONYMS

SURE SEEMINGLY UNRELATED REGRESSIONS

IMF INTERNATIONAL MONETARY FUND

FDI FOREIGN DIREVT INVESTMENT

ADF AUGMENTED DICKEY FULLER

GDP GROSS DOMESTIC PRODUCT

OLS ORDINARY LEAST SQUARES

GMM GENERALISED MOMENT METHODS

UNCTAD UNITED NATIONS CONFERENCE FOR

TRADE AND ADMINISTRATION

CLRM CLASSICAL LINEAR REGRESSION MODEL

CPI CONSUMER PRICE INDEX

VAT VALUE ADDED TAX

VIF VARIANCE INFLATION FACTOR

OECD ORGANISATION FOR ECONOMIC CO-

OPERATION AND DEVELOPMENT

ECOWAS ECONOMIC COMMUNITY FOR WEST

AFRICAN STATES

GLS GENERALISED LEAST SQUARES

AFRODAD AFRICAN FORUM AND NETWORK ON DEBT

AND DEVELOPMENT

ZIMSTATS ZIMBABWE NATIONAL STATISTICS

AGENCY

xiii
 LIST OF TABLES
TABLE 4.1 DESCRIPTIVE STATISTICS 53
TABLE 4. 2 HYPOTHESES TESTING SUMMARY 54
TABLE 4.3 SUMMARY OF HYPOTHESES TEST RESULTS 54
TABLE 4.4 UNIT ROOT TEST RESULTS 56
TABLE 4.5 HETEROSCEDASTICITY TEST RESULTS: BREUSCH-PAGAN- 57
GODFREY TEST
TABLE 4.6 MULTICOLLINEARITY TEST RESULTS: VIF 58
TABLE 4.7 MULTICOLLINEARITY TEST RESULTS: CORRELATION 58
MATRIX
TABLE 4.8 AUTOCORRELATION TEST RESULTS: BREUSCH-GODFREY 59
SERIAL CORRELATION LM TEST
TABLE 4.9 STABILITY DIAGNOSTICS TEST RESULTS: CHOW 61
BREAKPOINT TEST
TABLE 4:10 SUMMARY OF REGRESSION RESULTS 62

xiv
 LIST OF APPENDICES
APPENDIX 1: UNIT ROOT TEST RESULTS 86
APPENDIX 2: HETEROSCEDASTICITY TEST RESULTS 92
APPENDIX 3: MULTICOLLINEARITY TEST RESULTS 92
APPENDIX 4: AUTOCORRELATION TEST RESULTS 93
APPENDIX 5: STABILITY TEST RESULTS 94
APPENDIX 6: REGRESSION SUMMARY EQUATIONS AND DURBIN WATSON 95
RESULTS
APPENDIX 7: WALD RESTRICTIVE COEFFICIENTS FOR HYPOTHESES 98
TEST
APPENDIX 8: DESCRIPTIVE AND NORMALITY TEST RESULTS 100
APPENDIX 9: REGRESSION OUTPUT GRAPHS FOR VARIABLES 101
RELATIONSHIPS
APPENDIX 10: RAW DATA 102

xv
 LIST OF FIGURES
FIGURE 4.1 STABILITY DIAGNOSTICS TEST RESULTS: CUMULATIVE SUM 60
OF RECURSIVE RESIDUALS GRAPH

FIGURE 4.2 GRAPH ILLUSTRATING THE RELATIONSHIP BETWEEN 66

INCOME TAX AND ECONOMIC GROWTH

FIGURE 4.3 GRAPH DEPICTING THE RELATIONSHIP BEWTEEN VAT AND 67

ECONOMIC GROWTH

FIGURE 4.4 GRAPH SHOWING THE RELATIONSHIP BETWEEN INTEREST 68

RATE AND ECONOMIC GROWTH

xvi
 CHAPTER I

INTRODUCTION

1.0 Introduction

This chapter deals with the background of the subject matter under study. It also provides
a detailed explanation of the research problem which motivated the researcher to embark
on this study with the main view of proffering solutions. The chapter also goes on to look
at the objectives of the study, research questions, research hypothesis, significance of the
study, assumptions of the study, delimitations of the study, limitations of the study,
definition of terms and a summary of the chapter.

1.1 Background to the study

The utilisation of tax revenue in stimulating economic growth has been in play for several
decades across the globe. Taxation refers to compulsory levies imposed by authorities on
both corporations and individuals with a primary goal of raising revenue to finance
government expenditures. According to Weier (2006) quoting from Latham CJ in the
case of Mathews versus Chicory Marketing Board (Victoria) in 1938, a tax was explained
as an involuntary exaction in the form of money by public authorities for public purposes
which must be enforceable at law. However, in the recent times, the purpose of taxation
seems to have evolved and shifted to other objectives like resource allocation, income
redistribution and economic stability (Musgrave and Musgrave, 1989). Taxes are either
classified as indirect or direct taxes according to public finance literature. A direct tax is
one which cannot be shifted to another person by the taxpayer, thus both the incidence
and burden falls on the same person while an indirect tax can be shifted to another
person.

The world over, taxation remains a pinnacle of economic growth as the mandate of
governments of African nations have been constantly rising as a result of an upsurge in
the population size and deterioration of infrastructure (Abomaye-Nimenibo et al. 2018).
To that effect, a myriad of studies has been carried out in both developed and developing
countries producing a positive, negative and insignificant relationship between taxation

1
and economic growth. Riba (2016) found out that taxation has a positive relationship with
economic growth while Gale et al. (2015) found no relationship between taxation and
economic growth. On the other hand, McBride (2012) posits that there is a negative
significant relationship between taxation and economic growth.

In that respect, taxation has been utilised in stimulating economic growth in developed
countries like Turkey and United States of America. A study carried out by Korkmaz et
al. (2019) fortifies the importance of taxation in shaping economic growth. The study
revealed that indirect taxes have a positive and significant impact on economic growth
while direct taxes have a negative and significant impact on economic growth. The main
aim of this study was to investigate the impact of both direct and indirect taxes on the
growth of the Turkish economy using autoregressive distributed lad approach. On the
other hand, taxes also proved to be useful to the growth of the American economy. Kalas
et al (2017) looked at the impact of taxes on the economic growth of the United States of
America for the time period 1996 to 2016 and found out that income tax does not
influence economic growth in the United States. Moreso, Bakari et al. (2020) investigated
the nexus among domestic investment, taxation and economic growth in Germany using
cointegration and vector error correction model analysis. The study revealed that tax
revenue from both corporate and domestic investment has a positive effect on economic
growth. The study also supports the idea that economic growth can affect taxation hence
this study produced a bi-directional relationship between taxation and economic growth.

Apart from that, taxation was also proved to be central in influencing the growth of
OECD countries like Sweden, Germany, Italy, Iceland, Canada, Netherlands, just to
mention a few. Milenkovic and Kalas (2017) looked at the impact of taxes on economic
growth in Organisation for Economic Co-operation and Development (OECD) for the
time period 2012 to 2016. This study used Gross Domestic Product (DGP) per Capita as a
proxy for economic growth and utilised personal income tax, corporate income tax, social
security contributions, taxes on goods and services and tax on property as independent
variables. Results of the study indicated that tax revenue growth, personal income tax and
tax on goods and services have a negative impact on GDP per capita while corporate

2
income tax, social security contributions and property tax positively affect economic
growth. Andrasic et al. (2018) conducted another research buttressing the impact of
taxation on economic growth for OECD countries for the time period 1996 to 2016.
However, this study used GDP as a proxy for economic growth and similar independent
variables to that of Milenkovic and Kalas (2017)‟s study was used. The results indicated
that the growth of OECD countries‟ economies is enhanced by increases in taxes.
Moreso, it was found that property tax positively affects economic growth while tax on
goods and services has harmful effects on economic growth.

Quite a number of studies were also carried out in developing countries and they showed
the importance of taxation in stimulating economic growth though to some extent, it was
also proved that taxation can dent economic growth. Empirical evidence from African
countries like South Africa, Tanzania, Kenya and Tanzania supports this notion. Dlahla
and Hlalefang (2018) explored the impact of taxation on the growth of the South African
economy using an autoregressive lag for the time period 1981 to 2016. The findings of
the study revealed that economic growth, trade and openness, capital and taxes are
cointegrated. This fiscal policy was found to be very crucial in stimulating economic
growth. Kadenge (2021) also looked at the economy of Kenya and found a positive
correlation between taxation and economic performance. Ihenyen and Mieseigha (2014)
investigated taxation as an instrument of economic growth for the Nigerian economy and
found out that value added tax and corporate tax influence economic growth. Maganya
(2020) investigated the effect of tax revenue on economic growth in Tanzania and results
indicated that income tax is negatively related to economic growth while VAT is
positively related to economic growth.

Moreso, a study by Babatunde et al. (2017) is a testimony that taxation positively

influences the economic growth of African countries. This study supports the idea of
levying low tax rates affordable by many in African states in order to increase economic
growth. West African countries like Nigeria, Ghana, Sierra Leone and Burkina Faso also
utilise taxation in influencing the growth rate and patterns of their economies as
supported by Oboh et al. (2018). This study found out that taxation has a positive

3
 significant effect on economic growth for ECOWAS countries. An indication that
taxation is a viable economic driver for most African states as posited by Babatunde et al.
(2017) in his study of the impact of taxation on the economic growth of African
countries.

Zimbabwe is no exception to this notion and this has seen the Zimbabwean government
coming up with several tax reforms. This has seen the Zimbabwean government
establishing Zimbabwe Revenue Authority (ZIMRA) in 2001. The major goal was to
foster tax revenue administration ameliorations with the view of enabling tax revenue
collection which can then be tailored towards economic growth. However, Von soest
(2009) avers that there is scanty attention given to taxation studies especially in
developing countries and Africa in general. The scholar further lamented that researches
on tax system and administration in Africa is rarely sufficient and this area of study has
for long been neglected despite its great contribution in propelling economic
development. Babatunde et al. (2016) is also of the view that the relationship between
taxation and economic growth is inconclusive, complex and contentious. This notable
research inactivity in taxation and summonable varying findings on the effect of taxation
on economic growth motivates this study.

1.2 Statement of the problem

Like other countries, the Zimbabwean government charges taxation on economic

activities in the country providing themselves with enough financial resources to fund the
provision of public goods and services in the country. Many a times ZIMRA manages to
meet its target in revenue collection but economic growth remains low and continues on a
free fall. Capitalists believe that the purpose of taxation is not only fiscal but also non-
fiscal (Nimeni, 2015). According to Nimeni (2015), modern economics has made it
feasible to use taxes as a financial instrument for more efficient and flexible
implementation of multiple goals of economic and state funds. Taxes can also be used in
either contractionary or expansionary fiscal policy with the view of obtaining economic
stability. It therefore becomes imperative to find out why the Zimbabwean economy is
not responding to robust tax reforms being implemented by authorities.

4
 Moreso, ZIMRA exceeded its annual revenue collection by 24.65 percent and 5.86
percent for 2020 and 2019 respectively but economic growth as measured by real GDP
growth fell by 7.47 percent in 2019 and by 6.25 percent in 2020 though in 2021 a positive
growth rate of 5.6 percent was realised (World Bank, 2022). It is against this background
that this study is being carried out so that the effect of taxation on economic growth is
empirically tested and confirmed in a Zimbabwean context with the view of ascertaining
whether taxation has an effect on economic growth.

1.3 Justification of the study

Most of the theoretical and empirical literature on the subject matter shows different and
disaggregated results indicating that scholars did not reach an agreement on the effect and
nature of relationship between tax revenue and economic development. Different authors
have established a positive relationship between various types of taxes and economic
growth while others have established otherwise. While some authors established a weak
relationship, others established a strong relationship. Some authors established a
relationship only in the long run while others established both short-run and long-run
relationships. Moreso, most of the previous studies were undertaken in developed
countries where tax infrastructure and administration is well established compared to
developing countries like Zimbabwe. This therefore has motivated the researcher to find
out if the effect of taxation on economic growth in developing countries produces
divergent results with the main goal of proffering solutions that will improve growth of
upcoming economies.
1.4 General objectives

The main objective of this study was to find out the effect of taxation on economic
growth in Zimbabwe for the time period 2000 – 2021 using time series data with a view
of prescribing best measures that the Zimbabwean government can adopt when
implementing tax policies.

1.4.1 Specific Objectives

The study sought to achieve the following specific objectives:

5
 i. To examine the extent to which income tax revenue affects economic growth in
Zimbabwe over the time period 2000 – 2021.
ii. To critically evaluate the extent to which value added tax influences economic
growth in Zimbabwe over the time period 2000 – 2021.
iii. To assess the extent to which capital gains tax influences economic growth in
Zimbabwe over the time period 2000 – 2021.
iv. To investigate the other factors that has an impact on economic growth in
Zimbabwe.
1.5 Research Questions

i. Is there any relationship between income tax revenue and economic growth in
case of the Zimbabwean economy?
ii. Is there any relationship between value added tax and economic growth in case of
the Zimbabwean economy?
iii. Is there any relationship between capital gains tax and economic growth in case of
the Zimbabwean economy?
iv. Is there any relationship between interest rate and economic growth in case of the
Zimbabwean economy?
v. Is there any relationship between FDI and economic growth in case of the
Zimbabwean economy?
vi. Is there any relationship between inflation and economic growth in case of the
Zimbabwean economy?
1.6 Research Hypotheses

The study tested the following hypotheses in order to analyse the effect of taxation on
economic growth in Zimbabwe over the time period 2000 – 2021:

Hypothesis 1

H0: There is no significant relationship between taxation and economic growth in

Zimbabwe.

H1: There is a significant relationship between taxation and economic growth in

Zimbabwe.

6
 Hypothesis 2

H0: There is no significant relationship between macro-economic factors and economic

growth in Zimbabwe.

H1: There is a significant relationship between macro-economic factors and economic

growth in Zimbabwe.

Hypothesis 3

H0: There is no significant relationship between Foreign Direct Investment and economic
growth in Zimbabwe.

H1: There is a significant relationship between Foreign Direct Investment and economic
growth in Zimbabwe.

1.7 Significance of the study

1.7.1 To the student

The study assisted the learner to gain an in-depth understanding of how taxation and
other macro-economic factors influence economic growth. This may also act as a
motivation on part of the learner in complying with tax regulations of the country after
gaining an understanding of its significance to the growth of the economy.

1.7.2 To Great Zimbabwe University

The study and its outcome may be used by other learners as a starting point in carrying
other researches in the area of taxation and economic growth. The study might also be
used as a reference by future researchers as it adds value to the existing body of literature
in the field of Accounting.

1.7.3 To the government and ZIMRA

The study may provide an intuition to government and ZIMRA on how taxation and
macro-economic factors influences economic growth with the main view of provoking
informed policy variations that promote economic growth.

7
1.8 Limitations of the Study
A number of challenges were faced whilst conducting the research. This incorporates the
unavailability of useful data in the favourable form, time constraints and many other
factors as articulated below:

1.8.1 Financial constraints

The researcher encountered difficulties in meeting the costs relevant in conducting the
research especially transport costs and costs for buying software packages such as Eviews
that was required in regressing the data. Given such a situation, the researcher had to
draw from personal savings which were meant for other purposes.

1.8.2 Time constraints

Generally, the timelines for submitting the chapters of the research for review were
difficult to meet and the time allocated for completing the research was limited. The
researcher then concentrated on secondary data only which was readily available in order
to complete the research in time. Moreso, the researcher had to commit self during public
holidays in order to meet deadlines.

1.9 Assumptions of the Study

In the process of conducting the research, certain assumptions were taken into
consideration and the following are the major assumptions made:

i. Low levels of economic growth being witnessed in the country are due to bad tax
policies.
ii. Data collected from various sources is reliable and accurate.
iii. Macro-economic fundamentals are only determined and comprised of inflation
and interest rates.
1.10 Delimitation of the study

The study is restricted to the analysis of the effect of taxation on economic growth in
Zimbabwe during the time period 2000 – 2021 as well as an analysis of macro-economic
factors like inflation and interest rates that influence economic growth in Zimbabwe with

8
 main aim of finding out whether taxation is positively or negatively affecting economic
growth in Zimbabwe. The study also goes on to analyse the effect of economic growth on
taxation hence the study also touches on the bi-directional relationship between taxation
and economic growth. The study makes use of annual time series data from 2000 – 2021
only. Information utilised in this study was obtained from ZIMSTATS, IMF and World
Bank.

1.11 Definition of terms

Taxation – According to Lymer and Oats (2014), a tax is a non-voluntary levy imposed
by public power on wealth or people, expenditure and income for which nothing is
directly received by the tax payer in return.

Income Tax – According to Maganya (2020), it is a tax that is contributed by business

organisations (corporate income tax) and individuals (personal income tax) depending on
the earnings or profits attained during a financial year.

Tax revenue - Refers to the income which government collects from charging taxes
(Okenwa and Chike, 2021).

Economic Growth – According to Abata (2014), economic growth refers to an

economy‟s output of goods and services which is sometimes referred to as the Gross
National Product growth.

1.12 Acronyms and Abbreviations

VAT – Value Added Tax

ZIMRA – Zimbabwe Revenue Authority

GMM – Generalised Moment Methods

OECD - Organisation for Economic Co-oporation and Development

SURE – Seemingly Unrelated Regressions

GDP – Gross Domestic Product

9
 GNP – Gross National Product

ADF – Augmented Dickey Fuller

1.13 Organisation of the study

The research is organised into five different chapters. The first chapter which has already
been presented is the introduction of the study. Chapter two follows and is on literature
review. Chapter two starts with conceptual framework of the research, followed by a
discussion of theoretical framework and lastly empirical review. Chapter three is the
research methodology. It provides a detailed discussion of research paradigm, research
design, model specification, justification of variables, data collection tools, sources and
the estimation procedures for collected data. Then there is chapter four which majors on
data presentation and analysis while lastly there is chapter five which presents on
findings, conclusions and recommendations.

1.14 Summary

Chapter one was on the introduction of the study. It looked at the background to the
study, statement of the problem, purpose of the study, objectives of the study, research
questions, hypotheses, significance of the study, assumptions of the study, limitations of
the study and finally organisation of the study. The next chapter looks at literature
review.

10
 CHAPTER II

LITERATURE REVIEW

2.0 Introduction

The previous chapter introduced the subject matter under study and this chapter focuses
on reviewing literature done on the effect of taxation on economic growth. It is further
sub-divided into three distinct sections; conceptual framework, theoretical framework and
empirical review of literature.

2.1 Conceptual Framework

2.1.1 Taxation

According to Omotoso (2001), a tax is a compulsory obligation levied on properties and

individuals by a public authority, and is supported by a government decree, regulation or
law regardless of the service the tax payer gets in return. Thus taxes are not levied on a
quid pro quo basis. A tax is imposed by the government on income, product or activity.
Taxes are either direct or indirect. A tax that is levied on corporate or personal income is
called a direct tax while if it is imposed on the price of a good or service, it then becomes
an indirect tax.

Similar views were echoed by James and Nobes (2013) as well as Lymer and Oats (2014)
who defined a tax as a non-voluntary levy imposed by public power on wealth or people,
expenditure and income for which nothing is directly received by the tax payer in return
while Ojo (2017) views it as science involving imposition of taxes on citizens of a
country. AFRODAD (2011) adds that taxation plays a crucial role of redistributing
wealth from the rich to the poor in a country as well as providing authorities with funding
they need to finance economic growth and development. They further argued that
taxation is responsible for creating an environment conducive for business to be
conducted as well as shaping the way in which government activities are conducted. On
the other hand, the Institute of Chartered Accountants of Nigeria (2006) defines a tax as a
mandatory contribution of money legally enforced on citizens of a country by its
authorities.

11
Apart from that, Anyaduba (2004) opines that a tax is an instrument used by authorities to
raise public funds. Pianna (2003) views it as a product stemming from the application of
tax rate on tax base. A well-structured tax system can assist governments of developing
states to prioritise public spending, come up with stable institutions and raise their
democratic accountability. Ogbonna and Appah (2012) is of the view that a tax is
financial charge or some other levy imposed on tax payers by the state to which default to
honour is punishable by law. Goode (1984) defined a tax as an involuntary payment to
government by companies and households while OECD (2016) believes that taxes are
compulsory, unrequited payments to general government. They further explained that
taxes are unrequited in the sense that benefits provided by the government to taxpayers
are not normally in proportion to their payments. Okafor (2012) adds that every tax must
be based on valid statute for it to be legitimate. On the other hand, Edame and Okoi
(2014) defined it as a financial charge levied upon a taxpayer by a government and is
usually taken as a major source of revenue for the government for the financing of
various public expenditures. Thus these views concur with those of Okenwa and Chike
(2021) who opined that taxation is a primary source of government income the world
over.

Moreso, Abomaye-Nimenibo et al. (2017) posit that a tax is an involuntary payment

made by inanimate and animate beings to authorities of a country either indirectly or
directly to finance a number of its programs and non-compliance is meted with suitable
penalty. He further added that taxation is therefore viewed as the transfer of resources
such as income from the private sector to the public sector for use in order to meet some
if not all of a nation‟s social, political and economic objectives such as provision social
amenities and public goods like street lights. Appah and Oyandonghan (2011) share
similar views and believe a tax is a charge imposed on economic agents or their wealth
by authorities in order to raise resources to finance public goods, social amenities and put
in place a state conducive for the wellbeing of the society.

Apart from that, Azubike (2009) argues that a tax is a key player in every social set-up
across the whole world. Similar views are also given by Abomaye-Nimenibo et al. (2018)
and he believes a tax a vital part of every country‟s investment and growth pattern.

12
 Olasatiyan and Samkey (2011) defines a tax as an involuntary charge levied by the
government on the income of individuals and businesses as specified by a public
authority‟s decree, acts or case laws regardless of the precise amount of services provided
to the taxpayer in reciprocation. Nnoruga et al. (2018) further added that a tax is an
involuntary but non-criminal charge by a public authority using its agents on the gains,
income or consumption of its citizens.

Several taxes are levied in Zimbabwe. Nyatanga (2016) notes that customs and excise
duty, income tax, pay as you earn, presumptive taxes, value added tax, road tolls, capital
gains tax and surtax are the most common types of taxes levied in Zimbabwe. All these
types of taxes are backed by relevant statutes making them legitimate and enforceable.
ZIMRA is mandated with assessing, collecting and accounting for tax revenue on behalf
of government through the Ministry of Finance (Nyatanga, 2016).

2.1.2.1 Income Tax

Income tax is a tax that is contributed by business organisations (corporate income tax)
and individuals (personal income tax) depending on the earnings or profits attained
during a financial year (Maganya, 2020). According to Abomaye-Nimenibo et al. (2018),
this tax was first introduced in Great Britain in 1911 by Patt as a temporary levy. It was
then disbanded in 1915 and only to reappear in 1942 upto now it is in use in almost all
countries across the globe. Kagan (2021) defines income tax as a category of tax that
authorities of a country levy on income created by firms and individuals within their
areas of jurisdictions. In Zimbabwe, income tax is regulated by the income tax Act
Chapter 23:06.

2.1.2.2 Value Added Tax

According to Abomaye-Nimenibo et al. (2018) value added tax is not submitted by user
directly to a tax authority but it is paid to a seller when taxable items or service is sold.
The seller then submits it to a tax authority. It originated in France in the late 1950s to
replace the turnover tax, though the initial version of the Act was actually adopted in
Brazil in 1967. According to Abomaye-Nimenibo et al. (2018), VAT is a multi-stage
broad-based tax sales tax. He further added that VAT is levied on the value of an item or

13
 service as an additional charge hence value added. He also added that VAT is a tax
imposed on a measure of the contribution to the manufacture of an item or service by a
particular manufacturer and the income that is obtained from that contribution.

Furthermore, Nnoruga et al. (2018) is also of the idea that VAT is the total value that a
business entity adds to an item or service using its factors of production such as land,
capital labour and entrepreneurial capacity. Apart from that ActionAid International
(2018) defined VAT as an indirect tax imposed on the value added by manufacturers,
suppliers and service providers at every stage in a supply chain. This type of a tax is
normally passed to consumers. According to the Institute of Chartered Accountants
Academy (2020), VAT is a consumption tax attached to an item or service‟s sales price
and it indicates a tax on the value added to an item or service throughout its
manufacturing process. In Zimbabwe, VAT is regulated by the Value Added Tax Act
Chapter 23:12 which was introduced in 2004 to replace the Sales tax regime.

2.1.2.3 Capital Gains Tax

According to Kadenge (2021), capital gains tax is a tax that is levied on gains on disposal
of capital properties such as buildings and other investments. According ActionAid
International (2018), capital gains tax is a tax charged on the increase in value of a
property when it is sold. The institute further added that the tax originated in the USA in
1913 and in the UK in 1965. Apart from that, ZIMRA (2020) defined capital gains tax as
a tax imposed on the capital of gain arising from the disposal of a specified asset.
Specified asset refers to immovable property like land and marketable securities like
shares. In Zimbabwe, capital gains tax is regulated by the Capital Gains Tax Act Chapter
23:01 which was first promulgated in 1981.

2.1.3 Tax Revenue

Tax revenue refers to the income which government collects from charging taxes
(Okenwa and Chike, 2021). Contrary to that, Sion (2019) defined tax revenue as proceeds
from taxes on income and profits, taxes charged on goods and services (VAT), social
security contributions, taxes on ownership, payroll taxes and transfer of property. Tax
revenue may be extracted from various sources such as individuals, public enterprises,

14
 foreign aid and trade royalties on natural resources (Moss, 2016). Total tax revenue as a
percentage of GDP indicates the share of a country‟s output that is collected by the
government through taxes (OECD, 2016). They further added that, it can be regarded as
one measure of the degree to which the government controls the economy‟s resources.

The tax burden is measured by taking the total tax revenues received as a percentage of
GDP and it represents the responsibility for paying a greater portion of taxes. This
indicator relates to government as a whole and is measured in million United States
Dollars and as a percentage of GDP (Krugman, 2012). It provides a detailed explanation
on revenue collections made from various items such as income tax, pay as you earn,
VAT, corporate tax, presumptive tax, customs and excise duty, just to mention a few. To
that effect, (Brautigam, 2019) believes that tax collections from both indirect and direct
tax constitute tax revenue of a nation. Magu (2010) shares similar views and is of the
opinion that government collect tax revenue from both direct and indirect taxes.

2.1.4 Economic Growth

Economic Growth refers to the annual rate of increases in total production or income in
an economy (Poulsonand Kaplan, 2008). In the same vein, Abata (2014) defined
economic growth as an economy‟s output of goods and services which is sometimes
referred to as the Gross National Product growth. Ayres and Warr (2006) share similar
views and they believe that economic growth entails the increase in goods and services
produced by a nation. They added that it represents an increase in the capacity of an
economy to produce goods and services as compared from one time period to the other.
Magu (2010) adds that it refers only to quantity of goods and services produced.

Economic growth can be measured in several terms. It can be measured in physical or

monetary terms. Economic Growth can be measured using GDP per capita. It can also be
adjusted for inflation and be measured in real terms or be measured unadjusted for
inflation in nominal terms. Contrary to such views, Illyas and Siddiqi (2010) believe
economic growth measures growth in monetary terms only and does not look at other
aspects of development. Economic growth is either positive or negative. Negative
economic growth occurs during periods of recessions and depressions when National

15
 Income falls while positive economic growth is associated with a boom. The best
alternative for measuring economic growth is GNP and this enables comparisons of
multiple countries as all the statistics of several countries can be quoted in one currency
basing on the Purchasing Power Parity.

The level of economic growth has remained low in Zimbabwe (World Bank, 2020). In
2020 real GDP growth rate for Zimbabwe was at -8 percent while in 2019 it was -7.4
percent (World Bank, 2020). This leaves a lot to be desired given that during the same
period, the country recorded a substantial increase in general government tax revenue as a
percentage of GDP from 14.39 percent to 16.27 percent between 2019 and 2020. GDP
refers to the total monetary value of goods and services produced within a country for a
specified period of time regardless of who owns the means of production while GNP
refers to the total monetary value of goods and services produced within a country during
a period of time regardless of geographical location. OECD (2009) defined GDP as a
standard measure of the value of final goods and services produced by a country during a
period. They further construed that the term gross means that no subtraction has been
effected for the depreciation of machinery, buildings and other capital products used in
production by the resident institutional units of the country. On the other hand, Goossens
et al (2007) defined GDP as the fair cash value of all formerly acknowledged ultimate
goods and services manufactured within an economy for a specified time period.

2.1.5 Inflation

The term inflation refers to a perpetual rise in the cost of goods and services in an
economy for a specified period of time. Similarly, Adaramola and Dada (2020) define
inflation as general sustained rise in the prices of goods and services which is above a
given standard. In that same vein, Laidler and Parkin (1975) explained inflation as a
process of continuously increasing prices or equivalently of incessantly declining value
of money. According to the US Congressional Research Service (2021), inflation is again
defined as a general rise in the price of goods and services across the economy or in other
words, a general decline in the value of money. The institute further adds that as inflation
occurs, economic agents can procure less goods and services with a similar amount of
money. It is measured using the Consumer Price Index.

16
 According to Anochiwa and Maduka (2015), there are various types of inflation among
them being demand-pull inflation which stems from an increase in aggregate demand
without a reciprocal surge in supply. There is also cost-push inflation which occurs when
a decline in supply brought about by an increase in the cost of production of the
commodity (Anochiwa and Maduka, 2015). The two scholars further explain that the
capability of monetary authorities to consolidate low levels of inflation would raise the
ability to quicken economic growth. There is also structural inflation which stems from
changes in monetary policy (Adaramola and Dada, 2020). According to literature, the
effect of inflation on economic growth is in several dimensions. Inflation can deter
economic growth on one hand and promote it on the other hand or even have no effect.

According to Fischer (1993), high inflation promotes economic inefficiencies and

reduces growth by lowering the level of investment as well as the rate of productivity.
Adding to that Kasidi and Mwakanemela (2013) posits that numerous studies have found
a negative relationship between inflation and economic growth. Mamo (2012) also
supports the idea that inflation has a negative effect on economic growth. Contrary to
such findings Chude and Chude (2015) posits that inflation has a positive influence on
economic growth. Such findings were also supported by Enejoh and Tsauni (2017) as
well as Anidiobu et al. (2018) while Shuaib et al. (2015) is of the idea that inflation has
no effect on economic growth.

Generally, inflation rates in Zimbabwe had been very high averaging 255.20 percent,
557.21 percent and 92.54 percent for the years 2019, 2020 and 2021 respectively
(O‟Neill, 2021). Such high figures have a potential to negatively affect growth trends of
the economy which has remained low in the country by discouraging lending and keeping
bank deposits very low at ZW$210.7 billion as of 2021 (RBZ, 2021). Such a situation is
not favourable for economic growth and calls for the immediate fine-tuning of macro-
economic fundamentals.

2.1.6 Interest rate

According to Faure (2014) interest rate refers to the recompense satisfied by a debtor to a
creditor for the use of money for over a specific period of time and is usually expressed

17
as a percentage. In simplest terms it is the price of money. In like manner, Banton (2021)
also defined interest rate as the amount a lender charges a borrower and is a percentage of
the principal. Glantz and Kissell (2014) further adds that interest rate can either be
variable or fixed and a fixed interest rate is one that does not vary over the life of the loan
or investment regardless of the changing economic situation while variable interest rate is
subject to change and is often secured to an underlying index such as a one-year treasury
bill rate. On the other hand, Bean (2017) defines interest rate as either the reward
received for postponing consumption or the cost of consuming when resources are not
available. Mushtaq and Siddiqui (2016) posit that interest rate is the price of borrowing or
profit on lending.

Drobyshevsky et al. (2017) notes that interest rates is a crucial measure of financial
markets performance, which has a cogent impact on the economy as a whole but the
mechanisms and direction of the relation between interest rates and economic activity are
highly intractable to investigate. They further concluded that the link between interest
rates and economic growth is negative. Tobin (1965) posits that a higher real yield on
money as an alternative asset to capital has a negative effect on demand for capital in the
medium term. Apart from that, it is crucial to note that when interest rates are increasing,
both consumers and businesses will reduce spending. This will bring about earnings to
drop and stock prices to fall. On the other hand, when interest rates have excessively
declined, consumers and businesses are bound to raise their spending, pushing stock
prices to advertently rise.

According to Seabury (2022), changes in interest rates can have both positive and
negative effects on the markets. Mushtaq and Siddiqui (2016) is of the idea that a rise in
interest rate forces economic agents to save more as the former results in an increase in
income. They further added that an increase in interest rate also increases the price of
capital, resulting in a fall in investment within an economy hence depressing economic
growth. Muhammad et al. (2013) consider investment to be a function of interest rate and
income and they argued that interest rate raises the cost of borrowing money and hence it
reduces investments which results in a decline in the growth of the economy. Tokuoko
(2012) also supports a negative relation between the real interest rate and corporate

18
 investment while Pattanaik et al. (2013) observe that a lower interest rate can raise
economic growth. Surprisingly, Salahuddin et al. (2009) posits that the interest rate is
insignificant with regard to growth.

Currently interest rate is at sixty percent in Zimbabwe (RBZ, 2022). Such high level of
cost of borrowing discourages investment and national output is expected to remain low
across the country. The Zimbabwean economy is heavily dominated by Small and
Medium Enterprises who cannot sustain such high borrowing costs yet they are the ones
contributing to the GDP of the nation. This calls for immediate policy interventions in
order to ensure economic growth is promoted by making capital available for small
businesses.

2.1.7 Foreign Direct Investment

According to IMF and OECD (2022), Foreign Direct Investment reflects the aim of
obtaining a lasting interest by a resident entity of one economy (direct investor) in an
enterprise that is resident in another economy (the direct investment enterprise). It is a
category of cross-border investment in which an investor resident in one economy creates
a lasting interest and a significant degree of influence over an enterprise resident in
another economy (OECD, 2008). They further added that FDI is a channel through which
technology is transferred between countries as well as advancing international trade by
providing foreign market access to local firms and can be a vital vehicle for economic
growth. On the other hand, UNCTAD (2021) defined FDI as a long-term relationship
investment which depicts a lasting interest and control by a resident entity in one
economy (foreign direct investor) in an enterprise resident in an economy other than that
of the foreign direct investor (foreign affiliate).

FDI influences economic growth in several ways, either indirectly or direct. OECD
(2008) argues that FDI can provide financial stability, promote economic growth and
enhance the wellbeing of societies. They further added that FDI is a crucial source of
capital for a range of host and home economies. Kulu et al. (2021) is of the idea that most
upcoming economies are using the attraction of FDI as a way to promote economic
growth. Okwu et al. (2020) posit that FDI has a positive significant effect on economic

19
 growth. Similarly, Melnyk et al. (2014) also found out that FDI has a positive impact on
economic growth. However, Alfaro (2003) found out that FDI in the primary sector
negatively influences economic growth while FDI in the manufacturing sector positively
influences economic growth. Yabi (2010) is of the view that FDI does not affect
economic growth all the time, that is to say, countries with high economic growth proved
to have a direct influence of FDI but this was not found in countries with low economic
growth, owing to the heterogeneity of countries.

According to World Bank (2020), Zimbabwe‟s FDI for 2020 was US$150 million, a
39.74 percent decline from 2019 while that of its counterpart, South Africa had a
staggering US$50.4 billion for the same period and US$41.15 billion in 2021. This shows
that Zimbabwe is not a good FDI destination and such low levels of FDI negatively
affects growth of the economy hence the need to create a conducive investment climate in
order to reinvigorate slumping economic growth.

2.1.8 Taxation and Economic Growth

The effect of taxation on economic growth is bi-directional (Ouma, 2019). That is

taxation influences economic growth while on the other hand economic growth also
influences taxation. However, most studies analysed the effect of taxation on economic
growth ignoring the effect of economic growth on taxation. When economic growth
increases, tax revenue is expected to increase because an increase in economic growth
creates employment increasing tax base of the country. Moreso, an increase in economic
growth results in a responsive increase in income levels and holding all other things
constant, tax revenue is expected to increase. Economic growth also results in better
living standards and production activities as business booms increasing the tax base of a
country. This also improves tax compliance as most people could now afford to pay
taxes leading to more tax revenue.

According to Abata (2014), a tax system is one of the most powerful tools that any
government can use to stimulate and control its social and economic development. Most
governments in the current century heavily rely on taxation. Palic et al (2017) believes
emergence of the endogenous growth theory ushered new possibilities of examining the

20
influence of of endogenous determinants such as taxation on economic growth. Moreso,
Alinaghi (2015) strongly argues that the effect of taxation on economic growth is
strongly determined by the structure of a country‟s tax system.

The Public Finance theory provides that taxes have a negative effect on economic growth
(Palic et al. 2017). This is so because in most cases, taxes usher distortions on the
behaviour of economic agents. Generally, taxes influence the behaviour of economic
agents on savings, spending and leisure (Kerner-Skreb, 1999). Any tax policy has a
tendency of discouraging new investment incentives and this distorts entrepreneurial
decisions as well as demotivating employees to work. When employees substitute work
with leisure, productivity drops leading to low levels of economic growth. In most cases
it has been proved that taxes results in economic inefficiency in form of deadweight loss.

When explaining the effect of taxation on economic growth macro-economic theory, two
models are commonly used. These are the Neo-classical growth theory and the
endogenous growth model. Barro (1990) as well as King and Rebelo (1990) came up
with the most commonly used growth model which assumes that public expenditure and
taxes can dictate both output level and steady growth of a nation (Alinaghi, 2015).
According to the endogenous growth model, the impact of fiscal policy growth is
determined by the level and structure of taxation (Belaney et al. 2001). The model further
provides that economic growth is a product of economic agents‟decisions in terms of
human capital and physical wealth accumulation (Kesner-Skreb, 1999). In modern
economics, the effect of taxation on economic growth has also been proved to be positive
as supported by several researchers. Gale and Samwick (2016), Macek (2014) as well as
Yi and Suyono (2014) found out that taxation positively influences economic growth.
This has made the effect of taxation on economic growth inconclusive as propounded by
Babatunde et al. (2016), hence embarking on this study.

According to Myles (2007), economic growth and taxation are linked through economic
agents‟ decisions. Optimal choices of individual economic agents are altered by changes
in the level of taxes and this affects the growth of the economy. Myles (2007) further
argues that the effect of a tax change upon growth rate of the output is a determinant of
two distinct components which are actions taken by economic agents and the effect this

21
 has on growth. Thus any change in tax provokes a corresponding action from a tax payer.
Action choice rests upon policy adopted and implemented by the government. This
enables a variation in tax on growth to be calculated by dividing the change in growth by
the change in taxation Myles (2007). The effect of such an observation is that economies
need not respond in like manner to taxation. For example, when there is a tax increase,
economic agents need not to similarly decrease their human capital. Thus the
consequences on growth may not necessarily be similar. Myles (2007) clearly argues that
this depends on whether a country is labour intensive or heavily relies on research and
development. A country with a growth rate reliant on human capital accumulation is
likely to feel the effect of a tax rise than the one relying on research and development
expenditure. Such a view separates the effect of taxation between developed and
developing countries. This makes it apparent that comprehension of the effect of taxation
requires comprehension of the components that constitutes the grant change in growth
(Myles, 2007). This clearly shows that tax elasticity of actions of economic agents is not
only what matters to tax authorities when designing tax reform policies. It also shows
why it is crucial to comprehend channels through which growth originates.

Takuma and Lyke (2015) posit that taxation can have both negative and positive effects
on economic growth. They further added that the negative effects are due to distortions in
choices and effects of discouragement factors inherent in taxes while positive effect is
due to expenditure financed by the taxation. Keynesian theorists believe that there is a
causal flow from taxation to growth; hence policy makers can rely on taxation as a fiscal
policy instrument to manipulate economic activities (Takuma and Lyke, 2015). However,
classical growth theorists believe there is no causal flow between taxation and growth
making taxation an ineffective tool for fiscal policy (Takuma and Lyke, 2015). This
clearly indicates that the debate on the effect of taxation on economic growth is not
conclusive hence embarking on this study.

2.2 Theoretical Literature

2.2.1 Benefit Received theory

The benefit received theory developed by Lindahl in 1960 is grounded in the voluntary
interchange assertion. It is pinned in the belief that there is interchange link between the

22
 state and tax payers. That is citizenry make financial contributions through taxes while
the state reciprocates through supplying goods and services for the society‟s benefit.
According to Richard and Peggy (1973), the benefit proposition was first put forward by
Henrik Stockholm school economics. In particular, it was put forward by Knut Wicksell
(1896) and Erik Lindahl (1919). This was done with the main goal of evaluating tax
efficiency and fiscal policy.

This theory provides that an individual ought to be taxed in tandem with benefits s/he
ought to get from the state in form of provision of goods and services. The idea is that
members of the society should partake in the provision of their daily needs by the state.
This therefore entails that authorities should endeavour to provide services that match the
contribution of its people. Government is mandated to the provision of certain goods and
services to its people as compensation for taxes paid by them. In actual fact, a
proportional link between taxes contributed and benefit obtained from the taxes should
exist. Thus the theory presumes a condition of equilibrium between the marginal rate of
tax and marginal benefit received to decide the extent of taxes to be contributed (Bhartia
2009). In effect, a person that benefits more from the programs of the state, should as
well contribute more to the government, thus a „quid pro quo‟ is expected to subsist
though in actual fact this does not practically exist. The main impossibility to this
proposition is due to the difficulty in ascertaining the amount of government benefits,
such as police protection received by each tax payer.

2.2.2 Optimal Tax Theory

The optimal tax theory provides that a tax system must be selected to maximize a social
wellbeing function subject to a set of constraints. The theory touches on a number of
economic resolutions such as on whether the state should levy income or commodity
taxes (Gentry, 1999). It also expands to the idea of how tax rates should differ across
commodities as well as the progressiveness of a tax system. This has created a long
standing debate on the choice between direct (income) tax and indirect (commodity tax).
According to Gentry (1999), optimal tax theory takes into consideration a number of
particular issues of a tax system. These models share three common characteristics

23
according Gentry (1999). One, each model provides a collection of probable taxes for the
state, such as commodity taxes and the government‟s revenue needs. The models do
away with lump-sum taxes which would cause no economic distortion. Two, the models
apparently specify how economic agents respond to taxes. Three, the government has an
objective function for evaluating different tax configurations. In the simplest models, the
state‟s main objective is to restrict tax burden impact on tax payers while maximizing tax
revenue collections.

Optimal tax configuration of commodity tax rates is one of the most common strands of
the optimal tax theory. According to Gentry (1999), the main discussion is on whether
levying uniform commodity tax rates is optimal or not and this is now commonly referred
to as the Ramsey problem after the solution proposed by Frank Ramsey in his 1972
article. However, the answer to that is abstracting from the collection costs of
administering several tax rates; uniform tax rates are rarely optimal. Ramsey problem is a
static model with a representative consumer or alternatively it can have several
consumers with the same demand functions (Gentry, 1999). The government‟s objective
is to raise a certain amount of tax revenue while simultaneously restricting distortions
created by the tax system. Uniform commodity taxes are the best because they raise all
prices by the same magnitude hence not distorting prices of different goods.

Atkinson, (1977) provides that optimal tax theory is a subdivision of economics that
observes how taxes can be organized to obtain the slightest deadweight costs or to
achieve the finest result with regard to social wellbeing. Optimal tax theory is a discipline
of scheming and applying a tax that promotes efficiency and lowers distortion in the
market under given economic constraints. Generally, this principle comprises of
individuals' utility and the maximization problem entails the reduction of distortions
produced by taxation. A bland tax is a hypothetical tax, which eschew distortion and
promote efficiency. Holding all other things constant, taxpayers elect investment projects
with a lower tax return or one that is subject to exemption as compared to the one with a
higher tax return. Thus, economists are generally in agreement to the long standing fact
that taxes distort behaviour.

24
2.2.3 The expediency Theory of Taxation
The expediency theory asserts that every tax revenue collection proposal must pass the
test of practicability, which must be the only consideration when a state is selecting a
revenue collection proposal (Urio, 2020). This therefore entails that the social and
economic objectives of the society are not relevant basing on the idea that it is not
beneficial to come up with a tax which cannot be levied and collected efficiently.
However, there are pressures from economic, social and political groups (Urio, 2020).
Each group pushes for the prioritization of its own interests and authorities are always
pressured to redesign tax structure to take into consideration such interests (Bhartia,
2009). Moreso, the administrative set up may not be efficient to collect the tax at a
reasonable cost of collection (Urio, 2020).

However as already highlighted, an entire tax system cannot be built on the basis of
expediency as this may put the whole economy into jeopardy by overlooking crucial
areas of an economy that can be addressed through taxation. According to Bhartia (2009),
taxation offers a strong set of policy tools to authorities and must therefore be fully
utilized for correcting political, economic and social pitfalls of the society. Taxation can
be used to reduce inflation, unemployment, redistribute income from the rich to the poor
and redirect resources from unproductive sectors of the economy to productive sectors.

Practicability is a crucial tenet of every tax system. It would be useless to levy a tax that
is not collectable as it defeats the whole purpose of revenue generation. However,
whenever faced with a combination of distinct practicable taxes and distinct practicable
rates, an option must be done taking into consideration their likely consequences on
operation of the economy as a whole. In its entirety this proposition is not capable of
assisting government in making decisions on various practicable taxes and distinct
practicable rates. It is common knowledge for those suffering from the imposition of a
new tax or from a modification in an already set tax rate to recommend that the tax under
contemplation is not feasible. The expediency theory is relevant to the current study
because it seeks to explain several tax aspects capable of influencing economic growth of

25
 a nation that authorities must take into consideration when setting tax rates and designing
a tax system.

2.2.4 Ability-to-pay theory

The ability-to-pay philosophy of taxation posits that taxes must be imposed in line with
taxpayer‟s capacity to pay (Investopedia, 2020). The theory was espoused by Adam
Smith in 1776. The belief is that economic agents with higher incomes are able to pay
more in taxes and as such should contribute more than the poor. The idea is that wealth
people should be prepared to plough back a little to society which assisted them to reach
such greater heights. Advocates of the theory assert that a dollar denotes not a thing to the
affluent than a salaried person hence the affluent should be prepared to part ways with
such a little. The argument is that tax burden should be distributed among economic
agents in accordance with their capacity to bear it, taking into consideration all of the
relevant personal features and this makes personal levies the most suitable taxes. Thus
historically it was agreed that income is the best indicator of ability to pay.

Ability to pay theory gave birth to the progressive income tax concept and does not
consider the amount of services that taxpayers use. For example, all workers contribute to
AIDS levy despite not benefiting from the tax arrangement. According to Utz (2002), this
has made some theorists to reject the ability to pay proposition in its early stages, for
example, Knut Wicksell in 1986. Several English theorists like John Locke and Thomas
Hobbes also dissent the ability to pay hypothesis and they assert that equity should be
grounded on consumption rather than earnings. Indirect taxes such as VAT, excise duty
and sales tax can also be fairly reconciled to the ability to pay principle. This can be done
by instituting tax exemptions on necessities such as basic commodities like salt or by
metamorphosing tax rates in accordance with exigency of need. However, such policies
are overally ineffective and they contort consumer buying patterns and their intricacy
often makes them arduous to install. Thus the ability to pay theory is one serious strive by
governments to utilise tax policy to lessen inequality that can result in grave costly
distortions. The ability to pay theory therefore helps us to understand that when analysing
the effect of taxation on economic growth, it is crucial to first have an appreciation of the

26
 tax structure as some taxes are weak to an extent that their influence on output and
growth cannot be unfelt.

2.3 Empirical Literature review

The impact of taxes on economic growth in OECD countries by Milenkovic and
Kalas (2017)
The research investigated the effects of taxation in OECD countries from 2012 to 2016.
In this study GDP per capita was used as a proxy for economic growth while personal
income tax, corporate income tax, social security contributions, tax revenue growth, taxes
on goods and services and taxes on property were used as independent variables. The
study made use of panel regression model in analysing data. The results indicated that tax
revenue growth, personal income tax and taxes on goods and services have a statistically
significant negative impact on economic growth. The study also showed that corporate
income tax, social security contributions and tax on property positively influences
economic growth. Regression results further revealed that a one percent increase in
personal income tax leads to a 0.04 percent decrease in economic growth while a one
percent increase in social security contributions promotes growth in GDP per capita by
0.05 percent.

Estimating the impact of taxes on economic growth in the United States of America
by Kalas et al. (2017)
The research empirically estimated the impact of taxes on economic growth for the
United States economy using time series data from 1996 – 2016. The study employed tax
revenue growth, corporate income tax, personal income tax and social security
contributions as independent variables. GDP per capita was again used as a proxy for
economic growth. Diagnostic tests were also conducted in order to prevent spurious
regression results. Results indicated that personal tax and social security contributions are
weakly related to the growth of the GDP of the United States of America. Regression
results also revealed that tax revenue growth and social security contributions have a
significant effect on economic growth while corporate and personal tax have an
insignificant effect on economic growth of the US economy. A one percent increase in
tax revenue growth and social security contributions enhanced economic growth by 0.3

27
and 2.05 percent respectively. Corporate income tax was found to be the only form of tax
which negatively impacted economic growth in the US economy though it was not
statistically significant. The results of the study further showed that corporate income tax
and tax revenue growth are positively correlated with GDP.

The nexus among domestic investment, taxation and economic growth in Germany:
Cointegration and Vector Error Correction Model Analysis by Bakari et al. (2020)
The study examined the effects of taxation on economic growth in Germany both
theoretically and empirically. The study was carried out for the time period 1972 – 2016
using time series data. Correlation analysis was used to estimate the relationship among
tax revenue, domestic investment and economic growth. Moreso, an augmented
production function was used statistical estimation. Results indicated that theoretical
models isolate several routes through which taxation can influence economic growth. The
study supports a bi-directional relationship between economic growth and taxation. It was
also revealed that corporate tax and domestic investment have a positive effect on
economic growth while on the other hand economic growth also affects taxation.
Surprisingly no long-run relationships were found among domestic investment, taxation
and economic growth.

The effect of progressive tax on economic growth: empirical evidence from

European OECD countries by Elshani and Ahmeti (2017)
The paper investigated the impact of taxation on economic growth in countries that apply
a progressive tax system. Personal income tax, corporate income tax, value added tax,
non-tax revenues, social security contributions, property tax, customs and excise duty
were the independent variables used in this study. The study made use of panel data
regression analysis. GDP per capita was used as the proxy for economic growth. Twenty
European OECD countries that make use of progressive tax system were selected for the
study to make a total of two hundred and sixty observations. Results indicated that
personal income tax has a negative effect on economic growth in European countries that
apply a progressive tax system. On the other hand, corporate income tax and value added
tax positively influences economic growth in European countries that makes of use of a

28
progressive tax. Moreso, customs and excise duty have a negative significant impact on
economic growth. Similar results were also obtainable for non-tax revenues which proved
to have a negative significant effect on economic growth. On the other hand, social
security contributions, property tax and other tax revenues in countries that apply a
progressive tax system do not have any influence on economic growth.

The impact of tax rates on economic growth of Iran in the years 1981 – 2010 by
Dehghan and Nonejad (2015)
The study examined the impact of taxation on economic growth in Iran for the time
period 1980 to 2010. The study used corporate tax rate, business tax rate, indirect tax
rate, annual population growth rate, inflation rate and degree of trade openness as
independent variables. Data was analysed using the autoregressive distributed lag
technique. Results indicated that corporate tax rate, business tax rate and indirect tax rate
have a negative significant effect on economic growth, thus taxation was found to be
hampering economic growth in Iran. Apart from that, results also indicated that annual
population growth rate and trade openness positively influences economic growth while
inflation rate has a negative impact on economic growth.

The relationship between taxes and economic growth: evidence from Serbia and
Croatia Kalas et al. (2018)
The study aimed at examining the empirical effects of taxes on economic growth using
time series data from 2007 – 2016. In order to identify the impact of tax forms on
economic growth and their relationships, the study made use panel regression. Gross
Domestic Product was used the dependent variable and proxy for economic growth while
corporate income tax, value added tax, social security contributions and excise duty were
the independent variables. Multiple regression model was adopted in data analysis.
Regression results showed that corporate income tax, value added tax and social security
contributions have a positive impact on Gross Domestic Product while excise duty
negatively influences Gross Domestic Product. Moreso, value added tax was found to be
the only variable with a statistically significant impact on economic growth in these two
countries.

29
Tax revenue and economic growth in developing country: an autoregressive
distribution lags by Maganya (2020) in Tanzania
The study investigated the effect of tax revenue on economic growth in a developing
country and in particular Tanzania. The research was conducted for the time period 1996
to 2019 using autoregressive distributed lag model. Taxes on domestic goods and
services, income tax, domestic investment, Foreign Direct Investment, inflation and real
GDP were included as independent variables. The results indicated that domestic goods
and services are positively related to Gross Domestic Product Growth. They therefore
recommended the government to aim at growing, nurturing and sustaining tax base to
positively stimulate economic development. They also conducted preliminary tests such
as stationary tests and pair-wise granger causality tests. Results prove that taxes on goods
and services are positively related to economic growth and are statistically significant at
one percent significance level. Income taxes were found to be negatively related to
economic growth at a five percent significance level. Pair-wise granger causality test
indicated that there is bi-directional granger causality between taxes on goods and
services and economic growth at one percent significance level. The researchers then
recommended governments to aim at continuously growing, nurturing and sustaining tax
base in order to positively drive economic growth.

Taxation as an instrument of economic growth (The Nigerian perspective) by

Ihenyen and Mieseigha (2014).
The study examined taxation as an instrument of economic growth in Nigeria using
annual time series data from 1980 – 2013. The research made use ordinary least squares
in analysing the relationship between taxation and economic growth. Corporate income
tax and value added tax were taken as independent variables while economic growth was
included as a dependent variable. The results showed that corporate income tax and value
added tax have a significant relationship with economic growth indicating that indeed
taxation is an instrument of economic growth in Nigeria. The researchers recommended
that governments should ensure additional measures are put in place to ensure tax payers

30
do not evade and avoid paying taxes to enhance proper redistribution of income in an
economy.

The relationship between tax and economic growth: A South African Perspective by
Riba (2016)
The study investigated the relationship between tax and economic growth in South Africa
using quarterly data from 2003 to 2016. Personal income tax, corporate income tax and
value added tax were employed as independent variables. The autoregressive distributed
lag cointegration technique was used in data analysis with the view of examining the
long-run relationship between economic growth and taxation in South Africa. Results
obtained from the study indicate the existence of a long-run equilibrium at tax type levels
only. Furthermore, results support a bi-directional relationship between economic growth
and value added tax. Apart from that, causality analysis results support demand-following
hypothesis by which economic growth stimulates taxation. The results also support a
positive relationship between economic growth and taxation.

Tax revenue and economic growth in emerging markets: Is financial development

relevant by Kunofiwa (2021)
The study investigated the impact of taxation on economic growth in emerging markets
for the time period 2008 – 2018 using generalized methods of moments, fixed effects,
pooled ordinary least squares and random effects. The study further examined whether
financial development is relevant in enhancing taxation‟s impact on economic growth in
emerging markets. The study used panel data only. The results indicated that tax revenue
have a significant positive impact on economic growth in emerging markets under
dynamic generalised moment methods and random effects whilst financial development
was proved to have a significant positive effect on economic growth under the dynamic
GMM and pooled ordinary least squares. The results of the study support the idea that
financial development is a channel through which taxation influences economic growth
in emerging markets.

Taxation revenue and economic growth in Africa by Babatunde et al. (2016)

31
The study examined the effect of taxation on economic growth in Africa for the time
period 2004 to 2013. The study carried out several preliminary tests like descriptive
statistics, stationary tests and heteroscedasticity test. Ordinary least squares, fixed effect
regression and random effect GLS were the main estimation tests conducted in this study.
Findings supported the idea that taxation positively influences economic growth in
Africa. According to the findings of this study, high and weak levels of taxation are
favourable to economic growth as vouched by the economic effect of Ibn Khaldun‟s
theory on taxation which supports that lower rate has a positive impact on economic
growth, work, output and economic performance.

Tax revenue and economic growth in selected ECOWAS countries, evidence from
Seemingly Unrelated Regression Estimate [Sure Model] by Oboh et al. (2018)

The study analysed the impact of tax revenue, direct and indirect tax on economic growth
of five ECOWAS countries which are Nigeria, Ghana, Sierra Leone, Benin and Burkina
Faso. The research was conducted for the time period 2000 – 2015. Real GDP growth
was used as a proxy for economic growth as well as a dependent variable in the study.
Total tax revenue, trade openness, population, indirect tax revenue, direct tax revenue and
Foreign Direct Investment were included as independent variables. Findings shows that
total tax revenue has a positive significant impact on economic growth while an increase
in direct taxes dampens economic growth. Indirect taxes were found to be positively
influencing economic growth.

Causal relationship between government tax revenue growth and economic growth:
A case of Zimbabwe by Canicio and Zachary (2014)
The study investigated the effects of government tax revenue on economic growth in
Zimbabwe for the time period 1980 – 2012 using quantitative research methods. The
study also investigated short-run and long-run relationship between tax revenue and
economic growth in Zimbabwe for the same time period. The study made use of Granger
causality test, Johansen cointegration test and vector error correction model to analyse
data in a bid to infer the relationships. It was found out that taxation affects the allocation

32
 of resources and often distorts economic growth both theoretically and empirically.
Results further indicated that there is an independent relationship between economic
growth and total government tax revenue with thirty percent speed of adjustment in the
short-run towards equilibrium level in the long-run. Thus, there is fiscal independence
between tax revenue and economic growth in Zimbabwe.

2.4 Research Gap

In a nutshell, empirical evidence from previous studies shows different and disaggregated
results indicating that scholars did not reach an agreement on the effect and nature of
relationship between tax revenue and economic development. Different authors have
established a positive relationship between various types of taxes and economic
development while others have established otherwise. While some authors established a
weak relationship, others established a strong relationship. Some authors established a
relationship only in the long run while others established both short-run and long-run
relationships. Most literature concentrated on developed countries with well-established
tax administration and infrastructure ignoring developing countries like Zimbabwe and
hence embarking on this study.

2.5 Summary
The chapter touched on the review of conceptual, theoretical and empirical literature on
the subject of the effect of taxation on economic growth for both fully established and
upcoming economies. Taxation, tax revenue, economic growth, inflation, interest rate,
FDI, VAT, capital gains tax and income tax has been explored under conceptual
framework with the main view of paving a deeper understanding of these terms in
relation to the growth of economies as well as analysing how each of these factors
influence economic growth. Some of the theories discussed include the ability to pay
theory, expediency theory of taxation, optimal tax theory and the benefit received theory.
The next chapter dwells on research methodology, clearly outlining the research tools and
instruments used in the study.

33
 CHAPTER III
RESEARCH METHODOLOGY
3.0 Introduction
The previous chapter dealt with literature review and this chapter majors on the activities
and procedures that were adopted by the researcher in collecting data from various
sources for this study. According to Dawson (2002), a research methodology furnishes a
structure or a blueprint for undertaking a research. In line with that concern, the chapter
appraises the various methods used in data gathering, presentation and examination plan
to analyse the effect of taxation on economic growth in Zimbabwe. It is generally agreed
that the credibility of study findings heavily rely on the aptness and reliability of
procedures employed in data gathering and as well as analysis process. The main
objective of this section is therefore to offer arguments for the techniques that the
researcher utilised in collection and treatment of the data in a bid to provide responses to
research questions and objectives of the study. Central to this section is development of
the economic growth model for Zimbabwe with a correct rationale of the variables
embraced therein.

3.1 Research Paradigm

The research made use of a quantitative research paradigm as it was believed to be the
most suitable one in examining the effect of taxation on economic growth for it enables
the quantification of the influence that each independent determinant has on economic
growth (dependent variable). Quantitative research entails explanation of the phenomena
by assembling numerical data which is then analysed using statistical methods (ALiaga
and Gunderson, 2000). Quantitative ways are best suitable in instances where phenomena
can be quantified, computed and expressed numerically. This suits situation at hand as
with variables such as inflation, interest rate, FDI, capital gains tax, VAT and income tax
being quantifiable by standard measure. This has enabled this study to adopt the realism
research philosophy since it is pinned on the concept of isolation of reality from human
intellect and judgements when explaining events. This philosophy rests on the
supposition of a scientific technique to the expansion of knowledge. Realism is split into
two groups which are the direct and critical realism. Direct realism affirms that what you

34
 observe is what you get while critical realism asserts that humans can experience the
sensations and images of the real world. However, critical realism has exceptions and it
argues that the sensations and images of the real world can be beguiling and resultantly
may not show real world situations and their evolvement resulting in deceiving inferences
and conclusions. Realism is an extension of epistemology which is akin to positivism in
that it presupposes a scientific approach to the inception of knowledge. Such a
presupposition underpins the gathering and comprehension of data in explaining real
world events which is the basis of this study.

3.2 Research design

A research design is a general plan that assists in answering research questions in a study
(Saunders, 2012). This study made use of the explanatory research design with the main
view of analysing the effect of taxation on economic growth in Zimbabwe. In a clearer
sense, the explanatory research strategy has been elected as a valuable way of analysing
scenarios or problems in order to interpret the connection between variables as it assists
in fortifying internal validity. The design is also usable in researches that seek to find out
causal relationship between two or more variables. Quantitative research design has also
been taken as the most appropriate because it integrates theoretical considerations with
empirical observations and centres on examination of numerical data.

Deductive approach was considered to be the most suitable because this study narrows on
use of data to test theories and eventually concluded with generalization of the
proposition and recommendations. Numerous stages of the deductive approach were
taken into consideration and of these include theory, hypothesis, data collection, findings,
confirmation or rejection of hypothesis and the emendation of the theories of taxation.
More specifically, the study started by reviewing theories of taxation with a goal of
establishing how these theories help shape tax systems as well as assessing the impact of
these taxes on economic growth. Furthermore, a case study design has also been taken in
as one of the design strategies in this study by way of empirical findings from former
studies conducted for other economies on the effect of taxation on economic growth.
Kothari (2012) stresses that a case study is an in-depth study rather than breath as it lays

35
 down more importance on complete analysis of a restricted number of scenarios or
events.

This study is largely a desktop study as it predominantly makes use of secondary data
which is data that is already there and gathered for some other intentions. The utilization
of quantitative data is objective and removes the partiality that comes into effect from the
use of personal opinions and judgements associated with qualitative research.
Quantitative research is normally abreast to positivism, mostly if employed with
prearranged and hugely structured data gathering procedures. Quantitative research
enables one to identify and describe variables in order to find out relationships among
them. Following the work of Akwe (2017), Kalas et al. (2018) and Andrasic et al. (2018),
the model to be adopted in this study incorporates income tax, capital gains tax, VAT,
inflation, interest rate and FDI as explanatory variables in the economic growth function.

3.3 Target Population and Sample Design

The research population comprises of all individuals and officially registered business
entities trading in Zimbabwe and paying taxes. This therefore entails that those business
entities and individuals not paying taxes either through tax avoidance or evasion are not
part of this study because they have no contribution towards tax revenue which forms the
fiscus of the country and is central to this study. The population also excludes illegal
activities that are not taxable. The sample is then made up of all items that fit in the
population hence tax revenues figures used in this study is a final product of the defined
population as a whole.

3.4 Data collection Methods used

Data utilised in this research was primarily collected from government publications, the
IMF and World Bank, only to mention but a few. These sources of data were the once
selected because they closely follow and evaluate economic variables and economic
growth trends of the Zimbabwean economy longitudinally on a continuous basis. Thus
data used in this study is purely secondary and is in tandem with previous researches on
the effects of taxation on economic growth. Studies by Maganya (2020) in Tanzania,
Ahmad et al. (2018) in Pakistan, Bakari et al. (2020) in Germany, Kadenge (2021) in

36
 Kenya and Phiri (2016) in South Africa all made use of secondary data from international
and national statistical organisations such as world bank and IMF. Sources of this kind
are credible basing on the fact that they collect and compile economic data for Zimbabwe
and other countries as well as publishing such data periodically. Collected data on all the
variables under investigation formed the premise for all computations of partial
regression coefficients (β), correlation coefficient (R), coefficient of determination (R2)
and the adjusted R2. The data utilized in this study is strictly on an annual basis. Such a
decision has been influenced by limited availability of data in the preferred form and the
need to do things differently as compared with prior researches on this topic. The study
period of twenty-one years is compatible with the regression method adopted which
requires a sizable number of observations taking into consideration the combination of
variables employed that suffice multiple regression assumptions.

3.5 Theoretical Framework

In order to investigate the effects of taxation on economic growth in Zimbabwe, a
theoretical model of the form written below was adopted. According to Geetanjali and
Venugopal (2017), the main two types of taxes that influence economic growth are direct
and indirect taxes. The two scholars further add that besides taxation (A) there are also
macro-economic factors (B) like inflation and interest rates that influence economic
growth. This enables the grouping of drivers of economic growth into two as shown in
the function below:
EG = f (A, B)……………………………………………………………………………(1)
Where „A‟ is a grouping of components of taxation that influence economic growth in
Zimbabwe such as value added tax (VAT), capital gains tax (CGT) and income tax
(INCT) and „B‟ is a collection of macroeconomic factors affecting economic growth in
Zimbabwe such as inflation (INFL), interest rate (INTR) and Foreign Direct Investment
(FDI). The result is a theoretical model with the form below:
EG = f (VAT, CGT, INCT, INFL, INTR, FDI)………………………………………(2)

3.6.1 Empirical Model to be estimated

3.6.2 Mathematical and Statistical Model Specification

37
The study made use of Classical Linear Regression Model (CLRM) Ordinary Least
Squares (OLS) technique in approximating the regression equation to examine the effects
of taxation on economic growth. It adopted the model constructed by Akwe (2017) as
well as Riba (2016), following the work of Acemoglu (2007). The writers used the model
to examine the impact of taxation on economic growth and they made use of VAT,
income tax, capital gains tax, inflation, social security contributions and unemployment
as independent variables. In devising a model for this research, some modifications were
effected. The two independent variables, unemployment and social security contributions
were dropped. The rest of the independent variables have been retained in the model to
enable the researcher to generate insights in analysing the effects of taxation on economic
growth. The study also took into consideration other macroeconomic factors that are
believed to be influencing economic growth in Zimbabwe such as inflation, FDI and
interest rate. These three factors will also be treated as independent variables. With all
these variables, it therefore follows that the economic growth function is of the form
below:
EG = α + β1X1 + β2X2 + β3X3 + β4X4 + β5X5+ β6X6 + ε………………………….(3)
Where EG = measure of economic growth (dependent variable)
α = constant term or the intercept
β1 = partial regression coefficient for value added tax
β2 = partial regression coefficient for income tax
β3 = partial regression coefficient for capital gains tax
β4 = partial regression coefficient for inflation
β5 = partial regression coefficient for interest rate
β6 = partial regression coefficient for FDI
X1 = value added tax (independent variable one)
X2 = income tax (independent variable two)
X3 = capital gains tax (independent variable three)
X4 = inflation (independent variable four)
X5 = interest rate (independent variable five)
X6 = foreign direct investment (independent variable six)

38
 ε = is the error term or random residual term. It comprises of two components which are
sampling errors and purely disturbance random error (other factors that influence
economic growth which has been excluded in the operating model).
The model therefore becomes:
EG = α + β1VAT1 + β2INCT2 + β3CGT3 + β4INFL4 + β5INT5+ β6FDI6 + ε……….(4)

3.7.1 Model Justification

For validity and reliability‟s sake, the following explanations seek to justify the relevance
of the model used in this study.
3.7.2 Form of Regression Model and Analysis
Values taken by prognosticators or regressors variables are obtained from the data set,
making the model non-parametric. Twenty-two data points per variable provided the
basis for estimation of the values of the regressors. Non-parametric means not taking into
consideration any suppositions as to the form or parameters of a frequency distribution.
According McCune (2006) non-parametric regression is a type of regression analysis in
which the regressors do not take prearranged form but is created according to details
extracted from the data. This is in sync with non-parametric regression which needs large
sample sizes compared to regression pinned on a parametric model simply the data must
produce the model structure and the model estimates. Despite the fact that there are many
types of taxes and other macroeconomic factors influencing economic growth in
Zimbabwe, only variables anticipated to have a significant effect on economic growth
according to literature and the economic situation prevailing in the country were deemed
fit for modelling by the investigator.

The research utilised regression and correlation analysis. In statistics, regression analysis
is a statistical procedure for estimating the relationship between variables. With reference
to task at hand, prior studies on the effect of taxation on economic growth draw mainly
from the procedure of multiple regression and correlation analysis. Regression analysis
incorporates several methods for modelling and analysing variables, when focus extends
to link between a dependent variable and a sizable number of independent variables.
Thus, regression analysis assists one to clearly comprehend how and by how much the

39
distinct value of a dependent variable adjusts when any one of the independent variables
is altered while all of the other independent variables remain constant.

In order to examine the robustness of partial correlation between taxation and economic
growth, the investigator incorporated six variables only in the regression model. These
variables are believed to be the drivers of economic growth in Zimbabwe. In that regard,
the variables are VAT, capital gains tax, income tax, inflation interest rate and FDI. What
this then entails is that, regression analysis can be utilised in inferring causal relationships
between regressors and dependent variables. However, considerable care must be taken
here as this can result in illusions or misleading relationships. Taking for example,
correlation does not mean causation. In actual fact, this special type of multiple
regression and ordinary least squares are parametric. This means that the regression
function is specified in terms of limited number of undisclosed parameters that are
approximated from data sets.

The portion EG = α + β1VAT1 + β2INCT2 + β3CGT3 + β4INFL4 + β5INT5+ β6FDI6 is

called the deterministic portion while ε is termed the stochastic portion of the model. The
random error term is incorporated in the model to indicate movements in the dependent
variable which are not explained by independent variables imputed in the model. In more
precise terms, the error term stands for other variables not included in the regression
equation. The logical basis is that there are other tax components and non-tax factors left
out that may have an effect on the economic growth of the Zimbabwean economy. Thus,
the equation was approximated using the ordinary least squares procedure that lessens the
sum of squared errors promulgated by Legendre in 1805.

In actual fact, regression techniques persist to be a realm of intense research. In modern

times, new techniques have been introduced for robust regression. This has enabled
regression to incorporate correlated responses like time series and growth arcs.
Regardless of these successes, a simple regression model incorporates the following
variables; undisclosed parameters designated as β which may stand for a vector or
operand, the explanatory variables (X) and the dependent variable (Y). Given such
variables, a regression model joins (Y) to a function of (X) and β(Y ≈F(X;β)). This model

40
 will then produce hindsight on the set of explanatory variables that are related to
economic growth.

3.7.3 Justification of Variables

3.7.3.1 Dependent variable
Economic growth proxied by GDP is the dependent variable in the model. It is the
variable being explained by explanatory variables. Thus variations in economic growth
should be explained by regressors by way of a fitted model. Gross Domestic Product
which is a measure of economic growth is the total monetary value of goods and services
produced in a country regardless of who owns the means of production for a specified
period of time. GDP is calculated using three main methods which are the output
approach (value added), income approach and the expenditure approach. The expenditure
approach sums consumption, investment, government expenditure and net exports
(exports minus imports). The income approach sums total national income, sales taxes,
depreciation and net foreign factor income while the Value added approach simply
subtracts value of intermediate consumption from gross value of output.

Theory suggests that economic growth is measured using a number of variables such as
GDP per capita, GDP in physical terms and GDP growth rates. GDP per capita was used
by Dehghan and Nonejad (2015) as well as Tosun and Abizadeh (2005). GDP growth
rate was used by Maganya (2020). However, this study made use of GDP in physical
terms as a proxy for economic growth. Several prior studies used GDP in physical terms
as a proxy for economic growth (Ihenyen and Mieseigha, 2014, Andrasic et al. 2018,
Akwe, 2017 and Kalas et al. 2017).

3.7.3.2 Value Added Tax

Value added tax is the tax charged on goods and services. It is one of the independent
variables used in this study and theory has it that taxation negatively affects economic
growth though empirical evidence from studies that has been undertaken has proved
otherwise. For example, Kalas et al. (2018) found out that VAT has a positive significant
impact on economic growth. VAT was included because it is one of the types of taxes
difficult to evade or avoid as it is levied on goods and services that everyone needs for

41
 survival though in some instances it is exempted on some basic food stuffs. It is on this
basis that the researcher believes; VAT will provide meaningful inferences concerning
the growth of the Zimbabwean economy as compared to other types of taxes like
presumptive tax which is easily dodged by citizens and are selective in some instances.

Several studies have used VAT as an independent variable in the investigation of the
effect of taxation on economic growth. VAT can be measured in three ways that is in
general physical terms using percentage changes in VAT and as a percentage of GDP. In
this study, VAT is measured in physical terms and this moves abreast with former
researches such as that of Ihenyen and Mieseigha (2014), Kadenge (2021) and Elshani
and Ahmeti (2017).

3.7.3.3 Income tax

Income tax is the amount levied on total earnings or profits of organisations and
individuals. Personal income tax and corporate income tax is what constitutes income tax.
Personal corporate tax is charged on individual earnings while corporate tax is charged
on the profits made by business organisations. Theory provides that income tax has a
negative effect on economic growth. However, empirical researches has provided other
angles and this is seen in the findings of Kalas et al. (2018) which showed that income
tax has a positive effect on economic growth.

Income tax is measured in general physical terms, using annual percentage changes or
alternatively as a percentage of GDP. This study measured income tax as a percentage of
GDP as highlighted in the studies by Kadenge (2021), Gale and Samwick (2016) as well
as Kalas et al. (2018).

3.7.3.4 Capital gains tax

According to Kadenge (2021), capital gains tax is a tax that is levied on gains on disposal
of capital properties such as buildings and other investments. According ActionAid
International (2018), capital gains tax is a tax charged on the increase in value of a
property when it is sold. Capital gains tax has a negative effect on economic growth
according to theory but empirical studies produced different results as shown by literature

42
 review. This study measured capital gains tax as a percentage of GDP as indicated by
Dladla and Hlalefang (2018). Capital gains tax has been included because it is the only
type of tax that ropes in the wealth of the deceased through inheritance and in most cases
such wealth lies idle which can be the reasons for the prevailing low levels of economic
growth currently prevailing in the country hence its inclusion in the model.

3.7.3.5 Inflation
According to Siklos (2000), inflation is the general rise in the cost of living which is
measured in terms of Consumer Price Index. Smyth (1994) has shown that inflation
impacts growth negatively in the USA, each one percentage point increase in acceleration
causing a reduction of 0.158 percent in growth. Consumer Price Index is the principle
gauge in the prices of goods and services. It reflects the annual percentage change in the
cost of goods and services produced in a given year expressed in the base year prices.
The CPI reported figures tend to affect economic growth in several ways. When inflation
is high bank deposits are expected to decline as high inflation discourages lending and
this reduces the capital outlay available for businesses to borrow in order to finance their
activities which are crucial in driving economic growth. Such a situation favours
borrowing as repayment will be done when the money has lost; value that is at a loss.

This proxy has been used in prior studies, for example by Andrasic et al. (2018). The rate
of inflation is measured by the annual percentage change of consumer price index. It
reflects the annual percentage in the cost to the average consumer of acquiring a basket of
goods and services that may be fixed or changed at yearly intervals. The base year was
not fixed in this study in order to avoid the effect of hyperinflation of 2008 period.

3.7.3.6 Foreign Direct Investment

FDI is a channel through which technology is transferred between countries as well as
advancing international trade by providing foreign market access to local firms and can
be a vital vehicle for economic growth (OECD, 2008). UNCTAD (2021) defined FDI as
a long-term relationship investment which depicts a lasting interest and control by a
resident entity in one economy (foreign direct investor) in an enterprise resident in an
economy other than that of the foreign direct investor (foreign affiliate).

43
 FDI influences economic growth in several ways, either indirectly or directly. OECD
(2008) argues that FDI can provide financial stability, promote economic growth and
enhance the wellbeing of societies. They further added that FDI is a crucial source of
capital for a range of host and home economies. Kulu et al. (2021) is of the idea that most
upcoming economies are using the attraction of FDI as a way to promote economic
growth. Okwu et al. (2020) posit that FDI has a positive significant effect on economic
growth. Similarly, Melnyk et al. (2014) also found out that FDI has a positive impact on
economic growth. However, Alfaro (2003) found out that FDI in the primary sector
negatively influences economic growth while FDI in the manufacturing sector positively
influences economic growth. Yabi (2010) is of the view that FDI does not affect
economic growth all the time; that is countries with high economic growth proved to
have a direct influence of FDI but this was not found in countries with low economic
growth, owing to the heterogeneity of countries. FDI is usually measured using private
capital inflows as a percentage of GDP. Kunofiwa (2010) used Net Foreign Direct
Investment as a percentage of GDP and this study measured annual FDI figures as a
percentage of GDP. It would be interesting to see how FDI influences economic growth
in the Zimbabwean context.

3.7.3.7 Interest rate

According to Faure (2014) interest rate refers to the recompense satisfied by a debtor to a
creditor for the use of money for over a specific period of time and is usually expressed
as a percentage. In simplest terms it is the price of money. In like manner, Banton (2021)
also defined interest rate as the amount a lender charges a borrower and is a percentage of
the principal.

According to Seabury (2022), changes in interest rates can have both positive and
negative effects on the markets. Mushtaq and Siddiqui (2016) are of the idea that a rise in
interest rate forces economic agents to save more as the former results in an increase in
income. They further added that an increase in interest rate also increases the price of
capital, resulting in a fall in investment within an economy hence depressing economic
growth. Muhammad et al. (2013) consider investment to be a function of interest rate and

44
 income and they argued that interest rate raises the cost of borrowing money and hence it
reduces investments which results in a decline in the growth of the economy. Tokuoko
(2012) also supports a negative relation between the real interest rate and corporate
investment while Pattanaik et al. (2013) observe that a lower interest rate can raise
economic growth. Surprisingly, Salahuddin et al. (2009) posits that the interest rate is
insignificant with regard to growth. Currently, the cost of borrowing is high in Zimbabwe
and as such it is also believed that the cost of borrowing is playing a huge role in
constricting economic growth trends hence the reason behind its inclusion in the model.

3.8 Methods of Data Analysis used

The study carried out descriptive analysis, hypothesis testing, correlation analysis and
regression using excel and E-views software in a bid to provide answers to the study‟s
research questions.

3.8.1 Descriptive Analysis

Fall (2006) described statistics as the development and implementation of techniques to
the gathering, examination and exposition of perceived data from planned investigations.
A number of descriptive measures were performed in this research to help interpret data
variability and appropriateness for regression. The main goal was to aid in enhancing
reliability and validity of data utilised in this research. Such descriptive measures are
imperative for assessment and interpretation of data gathered for use in this study. The
most common measures of central tendency that can be used in any research include the
mean, median, mode and standard deviation (Muchengetwa, 2005). However, in this
study analysis was restricted to grouped data in increasing the reliability of data collected
from different sources. Validity is the capacity of a research tool to compute what it is
supposed to compute and reliability is the research tool‟s capacity to constantly and
correctly measure the concept under study (Wood et al. 2006). To that effect, the
researcher has to show how validity and reliability was determined (Polit and Beck,
2006).

The study also tested the assumption of normal distribution of the data variables in this
segment. Normality tests are conducted in order to ascertain that the variables employed

45
 in the model are normally distributed and the Jacque-Bera test is common for normality
tests. The Jacque-Bera test makes use of the mean based coefficients of skewness and
kurtosis to scrutinize for the normality of the variables. The level of asymmetry is
computed by skewness using values between -3 and 3 while a value of zero shows a
symmetrical distribution. Kurtosis is a measure of heaviness of the distribution tails. The
Jacque-Bera statistic has to be insignificant and histogram bell-shaped when data is
normally distributed. More specifically, Jacque-Bera probability should be greater than
0.05 for normally distributed data. Violation of normality assumption results in a few
problems as it does not contribute to bias or incompetence in regression models. It is
important only in computation of p-values for significance testing but this is only an issue
when a sample size is small. Normality is tested under the null hypothesis that data is
normally distributed against the alternative hypothesis of no normality.

3.8.2 Hypotheses testing

A hypothesis test is required for each variable to see if the coefficient of each variable is
statistically significant. The hypotheses tests done are as stated in chapter one of this
study. The hypothesis of this study is restricted to the Zimbabwean economy for
companies and individuals that pay taxes. The study utilised the Wald tests restrictive
coefficients using E-views software package version 7. Tests were conducted at both 5
and 10 percent level of significance.

3.8.3 Diagnostic tests

Whenever conducting statistical analysis, it is necessary to take into consideration the
presumptions under which that analysis was undertaken and to examine which of those
presumptions may suffice for the data used. Estimation using Ordinary Least Squares
procedures is rooted on a sizable number of presumptions. Violation of the presumptions,
results in estimates that are inconsistent and biased. Consequently, this study carried out
diagnostic tests with the main view of scanning conditions where the presumptions had
been interrupted so as to bring about robust regression results.

3.8.3.1 Unit root test

46
 A unit root test is applied to scan the availability and form of non-stationarity in time
series data. A number of time series indicate a drift over time thus they do not cohere to
the specifications of a weaker stationary. If estimates are done using non-stationary
series, spurious regression results are obtained whereby coefficients appears statistically
significant while in reality there is no relationship that exists between variables. Panel-
based unit roots are more thorough than unit root based on individual time series. This
study made use of panel-based Fisher-Augmented Dickey Fuller test which assumes a
null of the presence of a unit root. When variables are cointegrated of different levels, the
OLS estimates of those variables may produce super consistent results with the sense of
collapsing the true values than if they were stationary. ADF is a measure of the presence
of a relationship between current and past values of variables. It is favoured compared to
other methods because it is robust to handle both first order and higher autoregressive
processes and it avoids spurious regression results which are common when estimating
data with a trend. The test is run under the null hypothesis that data has a unit root and an
alternative hypothesis that data is stationary. The decision criterion is that, ADF t-statistic
should be greater than all the critical values at one percent, five percent and ten percent
level of significance ignoring the sign of the figures in order for the data to be regarded as
stationary. If data is not stationary at first order, it is differenced so that it becomes
stationary because data should be reasonably stationary and not fluctuating.

3.8.1.2 Heteroscedasticity
According to Koutsoyianis (1977), heteroscedasticity is where the standard deviations of
a variable, monitored over a specific time period are non-constant. This means that
variance of errors varies across observations yet LRM requires disturbances to have a
constant variance independent of time. This may result in mistakes of inferences though
the consistency of the regression is not affected. Resultantly, F-statistic for the overall
significance of regression is unreliable introducing bias into estimators of the regression
coefficients. In checking the presence of heteroscedasticity, the study employed the
Breusch-Pagan Lagrange Multiplier test since it is sensitive to normality presumptions
and has the advantage of scanning any linear form heteroscedasticity. It also allows
residuals to be modelled as a function of its non-stochastic residuals (Gujarati, 2004). The

47
 test was performed at five percent level of significance with the null hypothesis that error
variances are not constant over time and an alternative hypothesis that error variances are
constant over time. The decision criterion is that if F-statistic probability is greater than
0.05, the null hypothesis that error variances are not constant over time is rejected.

3.8.1.3 Multicollinearity
This is a problem that occurs when two or more regressor variables in a simple or
multiple regression model are extremely correlated (Maddala, 1992). This implies that
one can be lineally forecasted from the others with a considerable degree of accuracy.
This phenomenon happens when predictors used in a regression model shares a common
trend over time. Severe multicollinearity is a phenomenon because it can raise the
variance of the coefficient estimates and make the estimates very sensitive to smaller
changes in the model. Resultantly, R2 will be high but the individual coefficients will be
having high standard errors, so that the regression appears good as a whole yet the
individual coefficients are not significant. Range of estimates for parameters will be very
wide resulting in acceptance of the null hypothesis and significance tests that gives
inappropriate conclusions. This further brings some difficulty in making sharp inferences.
A number of procedures are used to correct this problem but this study only made of the
correlation matrix and the Variance Inflation Factors (VIF). This is so because a
correlation matrix helps us to find the direction and strength of each relationship thereby
enabling one to narrow findings in future studies as needed to determine causation
experimentally. It also has advantages of predicting human behaviour and can be cost-
effective.

3.8.1.4 Autocorrelation
The presumption that the residuals are mutually independent can be violated giving rise
to what is known as serial autocorrelation. In time series data, the successive residuals
tend to be highly correlated. Autocorrelation can also be protracted to cross section data
where the residuals are correlated with those of the neighbouring units (Maddala, 1977).
Serial correlation causes OLS to underestimate the variances and standard errors of
coefficients. It therefore raises the fit of the model. Resultantly, a researcher may
conclude that a relationship is present while in actual fact the variables under study are

48
 absolutely not related in any way. In this research, the Durbin Watson test was used in
checking if autocorrelation is present and negatively affecting the model of economic
growth‟s inferential fitness. Apart from that, the study also made use of the Breusch-
Godfrey Serial Correlation LM. The test was conducted under the null hypothesis that
data is serially autocorrelated and an alternative hypothesis that residuals are not serially
autocorrelated. A Durbin Watson statistic around two is acceptable and for Breusch-
Godfrey Serial Correlation Lagrange Multiplier, the decision criteria is to reject the null
hypothesis when the F-statistic probability for all variables are greater 0.05, that is when
they are insignificant with large p-values at a significance level of five percent.

3.8.1.5 Stability Diagnostics

A process is regarded as stable when all the response parameters used in measuring the
process have both constant means and variances over time. It must also have a constant
distribution. The study utilised stability diagnostics with the view of individually
estimating subsamples and find out if there are any significant differences in the
estimated models. Significant variations show structural movements in the relationship of
variables under study and since the data passes through a highly volatile period of 2007-
2008 this was considered an imperative step in enhancing the long-run relationship
between variables under study. The study made us of the chow breakpoint test and the
CUSUM test. CUSUM test is based on the cumulative sum of the recursive residuals
(Brown et al. 1975). This option plots the cumulative sum together with the five percent
critical lines. The test finds parameter instability if the cumulative sum moves outside the
area between the two critical lines for both tests. Chow breakpoint takes significance
differences of estimated equations into consideration and a significant difference shows
structural changes in the relationship of variables under study. If p-values are significant,
it implies the instability of coefficients over time. The test was run under the null
hypothesis that data is not stable across regimes and an alternative hypothesis that data is
stable across regimes at a five percent significance level. The decision criteria for chow
breakpoint is to reject the null hypothesis if the F-statistic probability is greater than 0.05.

3.9 Regression analysis

49
 Regression analysis is the best technique to be utilised when an analysis encompasses
more than two variables. Regression has been done on the regressors and dependent
variable with the main view of finding out the effect of taxation on economic growth.
Discussions are restricted on the independent variables that significantly affect economic
growth in Zimbabwe. Dummy variables were also incorporated in running the regression
models.

3.10 Research Instruments

This study utilised Microsoft excel and E-views version 7 as data analysis tools. The
choice of the packages was based on versatility in terms of statistical functions. These
packages were used to analyse the effect of taxation on economic growth in Zimbabwe
for the time period 2000 - 2021. Excel was used to draw the graphs used for illustrating
relationships and trends over time. E-views was used to perform the regression models.
Excel was also used to formulate data sets of the variables for the study and a time series
data set became available. Apart from that data was also analysed for precision before
fitting it into the research tools.

3.11 Data Presentation and analysis plan

In a bid to generate meaningful and precise interpretations, the research presented data in
tabular form, charts, graphs, tables and in qualitative form. Tables were mainly used to
group and organize data so that it becomes manageable and easy to interpret. Moreso, the
use of tables was meant to promote the display of summarized information accurately and
concisely effective, although they are not as visually stimulating as pictorial forms of
presentation. Graphs were used because they are the most appropriate for clear
presentation of analysis of the impact of each variable on the growth of the Zimbabwean
economy and they clearly produce trends for easy analysis of variations in any given
variable over time.

3.12 Summary
The chapter majored on research methodology, clearly interpreting how secondary data
was gathered and assorted for regression analysis including the research tools utilised. To
ensure reliability and validity of both the model and secondary data, some diagnostic tests

50
were performed. These are the unit root test, autocorrelation test and heteroscedasticity
test among other crucial tests. Moreso, several descriptive statistics tests were also
conducted for the same purpose. This includes measures such as the mean, median and
standard deviation. The data collected from secondary data sources was analysed using
statistical tools such as excel and e-views. Data presentation was made in form of charts,
graphs and tables. A combination of data analysis and presentation forms the basis of the
preceding chapter.

51
 CHAPTER IV
DATA PRESENTATION, ANALYSIS AND DISCUSSION
4.0 Introduction
The previous chapter touched on research methodology and this chapter dwelt much on
unveiling and analysing research results. Regression results pertaining to economic
growth have been presented and analysed in different forms in this chapter. Some results
have been displayed in tabular form, followed by comprehensive explanations of the
results. Tests done on descriptive statistics and hypotheses testing has been dealt with as
part of data analysis aids. Moreso, diagnostics tests carried out on the model such as unit
root test, autocorrelation test, multicollinearity test and stability tests forms part of this
section. Finally, a summary chapter is given at the end of the section.

4.1 Descriptive statistics for economic growth and its influencing factors
Table 4.1 shown below is an insert revealing the descriptive statistics that forms part of
this piece of work comprising the mean, mode, median and standard deviation among
many other measures of dispersion. The minimum and maximum values have also been
included and they assist in checking outliers in the data used in the regressions done. The
number of observations is 22 for each and every variable included in this study. VAT has
a relatively low standard deviation of 1.08 and this indicates a low variability in VAT
data figures used for regression. FDI has the lowest standard deviation of 0.36 and this
shows a high degree of its reliability in its contribution towards explaining variations in
economic growth in Zimbabwe as compared with other countries in the same region.
GDP has the highest standard deviation of 416.24 and this reveals that there is great
variability in the data. Such information is imperative in ranking the usefulness of the
variables in explaining variability that takes place in the dependent variable from time to
time. This is as indicated in the results presented in table 4.1 below. The mean for the
dependent variable (economic growth) is 9.25 and this suggests that the Zimbabwean
economy is growing at a relatively fair growth rate when compared with other countries
in the same region with it. The mean interest rate is 41.46 percent which shows that
borrowing is expensive in Zimbabwe and such a scenario is likely to hamper economic
growth as most entrepreneurs will find it expensive to borrow. The rate of inflation is also

52
 too high at 47.91 percent making it unconducive for business entities to thrive as
economic agents with excess funds will be discouraged from lending keeping bank
deposits low in the country.
Table 4.1 Descriptive statistics for economic growth and its influencing factors
GDP INCT VAT CGT INFL INTR FDI
Mean 9.2522 27.9905 1.1509 3.7409 47.9077 41.4586 0.1955
Median 1021.00 30.4400 5.2708 8.0108 3.2550 39.7150 0.1100
Maximum 1465.00 34.1800 3.7009 2.3210 557.2100 60.6700 1.7200
Minimum 357.00 19.9800 1.4208 1.0708 -72.7300 30.7600 -0.0200
Std. Dev. 416.24 4.5001 1.0809 7.9409 134.7826 8.1960 0.3596
Skewness -0.03686 -0.3481 1.1347 2.1019 2.7479 0.9116 3.6713
Kurtosis 1.3129 1.6454 3.3112 5.4509 10.5367 3.1558 16.1563

Jarque- 2.6140 2.1264 4.8101 21.706 79.75639 3.0691 208.086

Bera 1 7
Probabilit 0.2706 0.3453 0.0903 0.0000 0.0000 0.2156 0.0000
y

Sum 20355.00 615.7900 2.5210 8.2310 1053.970 912.0900 4.3000

Sum Sq. 3638404.0 425.297 2.4419 1.3221 381493.4 1410.674 2.7149
Dev. 0 3

Observati 22 22 22 22 22 22 22
ons
Source: Eviews Software Package (Version 7)
From table 4.1 above, it can be concluded that FDI, inflation and capital gains tax are
normally distributed variables at a five percent significance level while Value Added tax
is significant at ten percent level of significance. This is so because their Jarque-Bera is
greater 0.05 while income tax, interest rate and GDP are not normally distributed as
evidenced by their p-values which are by far below 0.05. However, the residuals indicate
that data is normally distributed as witnessed by Jarque-Bera probability of 0.69 which is
by far greater than 0.05 when taken at a five percent level of significance and as shown
by the bell-shaped graph in appendix eight attached.

4.2 Hypothesis test results for economic growth and its influencing factors

53
 The hypothesis test conducted in this study is limited to the Zimbabwean economy and
all the companies that contribute tax to the government of Zimbabwe. The hypothesis
was carried out using restrictive coefficients for Wald tests at one percent, five percent
and ten percent level of significance.
Table 4.2 Hypotheses testing summary with GDP as independent variable.
Hypothesis Null Hypothesis Alternative Hypothesis
Hypothesis 1 There is no significant There is a significant
(Taxation) relationship between relationship between
taxation and economic taxation and economic
growth in Zimbabwe growth in Zimbabwe
Hypothesis 2 There is no significant There is a significant
(macro-economic factors) relationship between macro- relationship between macro-
economic factors and economic factors and
economic growth in economic growth in
Zimbabwe Zimbabwe
Hypothesis 3 There is no significant There is a significant
(FDI) relationship between FDI relationship between FDI
and economic growth in and economic growth in
Zimbabwe Zimbabwe

Table 4.3 Summary Results of Hypotheses Testing using GDP as the independent
variable.
Variable Coefficient Prob Decision Findings
Hypothesis 1 1.2832 0.0000 Reject H0 The effect of taxation on economic growth
Taxation is statistically significant and this means
that taxation is expected to positively
impact the growth of the Zimbabwean
economy.
Hypothesis 2 1.7917 0.0000 Reject H0 The effect of macro-economic factors on
Macro- economic growth is statistically significant

54
 economic and this therefore means that macro-
factors economic factors are expected to
positively influence the growth of the
Zimbabwean economy.
Hypothesis 3 8.5512 0.0000 Reject H0 The effect of FDI on economic growth is
FDI statistically significant and FDI is
therefore highly anticipated to positively
impact the growth of the Zimbabwean
economy.
Source: Eviews Software Package (Version 7)
4.3.0 Diagnostics Test Results for factors that influence the growth of the Zimbabwean
economy.
In order to prevent chances of obtaining and interpreting spurious regression results,
diagnostic tests were carried out. These results are shown in the tables that follow:

4.3.1 Unit root tests for factors influencing economic growth in Zimbabwe
The general rule is that econometric classical regression for time series data only hold
when the variables are stationary over time; that is cointegrated 1(0). If variables are
integrated of order 1(1) or higher, they fail to suffice the requirements of econometric
assumptions but if short-run and long-run relationships exist, certain combinations of 1(1)
are likely to be 1(0) hence amenable for OLS estimation. If this holds, the variables are
said to be co-integrated and the OLS estimates of such variables are superconsistent and
collapse their true values more quickly than if they were stationary. In testing the
presence of a unit root, the study utilized the panel based Augmented Dickey-Fuller test
and a summary of the results for all the variables are as shown in the table 4.4 below:
H0: Non-stationary (unit root)
H1: Stationary (no unit root)

55
 Table 4.4 Unit Root Test Results for economic growth and its determinants

Variable(s) ADF test Mackinnon Prob Order of Decision

value integration
GDP -6.8671 -3.8315 0.0000 1(2) Stationary @
1%
INCT -5.6544 -3.8085 0.0000 1(1) Stationary @
1%
VAT -4.5646 -3.8085 0.0020 1(1) Stationary @
1%
CGT -4.3724 -3.8085 0.0030 1(1) Stationary @
1%
INFL -5.4904 -3.8085 0.0030 1(1) Stationary @
1%
INTR -4.0133 -3.8085 0.0065 1(1) Stationary @
1%
FDI -4.4716 -3.7880 0.0022 1(1) Stationary @
1%
Source: Eviews Software Package (Version 7)
The decision to either accept or reject the null hypothesis is based on a comparison of the
ADF statistic and the critical values (McKinnon values) at one percent, five percent and
ten percent level of significance. If the ADF test statistic is greater than the critical values
at all levels, the decision is to reject the null hypothesis that variables under study has a
unit root. In this case, the null hypothesis is rejected in favour of the alternative
hypothesis for all the variables under study. The rest of the variables became stationary
after first differencing while GDP became stationary at lag two.

4.3.2 Heteroscedasticity Test Results for economic growth and influencing factors
The presence of heteroscedasticity in the residuals was checked using Breusch-Godfrey
test. This was done under the null hypothesis that residuals have heteroscedasticity. The

56
 decision criterion is that F-statistic probability should be insignificant, that is greater than
0.05 if residuals have homoscedasticity.
H0: Error variances are unequal or non-constant (heteroscedasticity)
H1: Error variances are equal or constant (homoscedasticity)
Table 4.5 Heteroscedasticity Test Results: Breusch-Pagan-Godfrey Test
F-statistic 0.8015 Prob. F(6,15) 0.5832
Obs*R-squared 0.3408 Prob. Chi-Square (6) 0.5009
Scaled Explained SS 3.1808 Prob. Chi-Square (6) 0.7858
Method: Least Squares Included variables = 22
Variable (s) Coefficient Std. Error T-statistic Prob
C 95999.73 48396.11 1.983625 0.0659
INCT -2225.648 1539.491 -1.445704 0.1688
VAT 1.3505 1.1505 1.177473 0.2574
CGT -1.4606 1.3306 -1.097225 0.2898
INFL 35.91259 51.14479 0.702175 0.4933
INTR -779.9395 1008.227 -0.773575 0.4512
FDI -3618.580 13373.21 -0.270584 0.7904
Source: Eviews Software Package (Version 7)
The results of Breusch-Pagan Godfrey Test show that the problem of heteroscedasticity
did not exist at a five percent level of significance because F-statistic probability (0.5832)
is far greater 0.05. Therefore, the null hypothesis is rejected in favour of the alternative
hypothesis, that is to say that standard deviations are constant over time.

4.3.3 Multicollinearity Tests Results for Economic Growth and influencing factors
This test was carried out under the null hypothesis that explanatory variables are
correlated over time (presence of multicollinearity) against an alternative hypothesis that
there is no correlation. In order to detect the presence of multicollinearity, the study made
use of the correlation matrix and variance inflation factor. The decision criterion for VIF
is that coefficients should not be greater than ten while for correlation; values should not
exceed 0.8 or -0.8.
H0: Presence of multicollinearity
H1: No multicollinearity

57
Table 4.6 Multicollinearity Test Results: Variance Inflation Factors
Variable (s) Coefficient Variance Uncentered VIF Centered VIF
INCT 55.3923 57.0325 32.7567
VAT 3.0915 9.6130 6.6543
CGT 7.7017 7.3183 4.7864
INFL 0.1149 2.8937 1.8906
INTR 25.2036 57.6324 48.0980
FDI 7365.469 1.5266 1.0987
Source: Eviews Software Package (Version 7)
Table 4.6 above indicates VIF coefficients for multicollinearity test and as a general rule all
values should not exceed 10 but in this case values for income tax and interest rate exceed ten.
This suggests the existence of the problem of multicollinearity in the two independent variables.
However, the two variables were retained as they are regarded important for the study.

Table 4.7 Multicollinearity Test Results: Pairwise Correlation Matrix for

correlation analysis on the factors that affects economic growth in Zimbabwe.
GDP INCT VAT CGT INFL INTR FDI
GDP
1.0000 0.8960 0.6626 0.2866 0.1083 0.3460 -0.1495
INCT
0.8960 1.0000 0.6634 0.3224 0.2032 0.5245 -0.0266
VAT
0.6626 0.6634 1.0000 0.8216 0.6119 0.8223 -0.2186
CGT
0.2866 0.3224 0.8216 1.0000 0.7632 0.7671 -0.1788
INFL
0.1083 0.2032 0.6119 0.7632 1.0000 0.6026 0.0117
INTR
0.3460 0.5245 0.8223 0.7671 0.6026 1.0000 -0.0243
FDI
-0.1495 -0.0266 -0.2186 -0.1788 0.0117 -0.0243 1.0000
Source: Eviews Software Package (Version 7)
Table 4.7 above shows correlation of the variables used in this study. The pairwise
correlation matrix shows the nature of the relationship as well as the degree of correlation
between variables used in this study. The results show that FDI is the only variable
negatively correlated with GDP while the rest of the variables are positively correlated
with GDP. Income tax has a highest correlation with GDP as evidenced by statistic of

58
 0.8960 while FDI has the lowest correlation with GDP as backed by a statistic of -0.0266.
A perfect correlation also exists between GDP and income tax. All the variables used in
this study pass the correlation test hence it can be concluded that the problem of
correlation did not exist among variables used as there is no variable with a statistic
greater 0.8 or more than -0.8 as shown in the table above.

4.3.4 Autocorrelation Test Results on economic growth and the factors that affects its
growth
In checking the problem of serial autocorrelation, the study made use of the Durbin
Watson statistic and the Breusch-Godfrey Serial Correlation LM test. This diagnostic test
was carried out under the null hypothesis that residuals of explanatory variables have
serial autocorrelation.
H0: r > 0 (Positive serial correlation)
H1: r = 0 (No Serial Autocorrelation)
If there is no serial autocorrelation, the Durbin Watson Statistic will be around two as
shown on the main regression equation. The Durbin Watson Statistic obtained in the main
equation run is 2.09 as indicated in Appendix six. Moreso, a look at the Breusch-Godfrey
Serial Correlation LM test in Table 4.8 below shows that the problem of serial
autocorrelation did not exist since the F-statistic for all the variables used in this study are
greater 0.05. The null hypothesis that residuals have positive serial autocorrelation is
therefore rejected.
Table 4.8 Autocorrelation Test Results: Breusch-Godfrey Serial Correlation LM
Test.
F-statistic 0.6074 Prob. F(2,14) 0.5585
Obs*R-squared 1.7542 Prob. Chi-Square (2) 0.4160
Test Equation: All variables included
Variable (s) Coefficient Std. Error T-statistic Prob
INCT
1.8388 7.8139 0.2353 0.8174
VAT
2.1708 6.0408 0.3598 0.7244
CGT
-9.1910 9.0409 -0.1016 0.9205

59
 INFL
-0.1008 0.3606 -0.2795 0.7839
INTR
-1.5010 5.3266 -0.2818 0.7822
FDI
-22.2306 90.7340 -0.2450 0.8100
RESD (-1)
-0.2881 0.2951 -0.9761 0.3456
RESD (-2)
-0.2586 0.3128 -0.8267 0.4223
Source: Eviews Software Package (Version 7)
4.3.5 Stability Diagnostics Test Results for Economic Growth and Factors Responsible for
its Growth in Zimbabwe
As a general rule, time series data have to possess parameters with constant means,
variances and a constant distribution over time in order for it to be termed stable. There
should be no structural changes in the relationship of variables used in a study for this to
hold. This study utilized the CUSUM test and the chow breakpoint test. This was done
under the null hypothesis that data is not stable over time against an alternative
hypothesis that data is stable over time.
H0: Data is unstable over time
H1: Data is stable over time

12

-4

-8

-12
06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21

CUSUM 5% Significance

Source: Eviews Software Package (Version 7)

60
 Figure 4.1 Stability Diagnostics Test Results: Cumulative Sum of Recursive
Residuals Graph.
Table 4.9 Stability Diagnostics Test Results: Chow Breakpoint Test
Chow Breakpoint Test: 2008
Null Hypothesis: No breaks at specified break points
Varying Regressors: All Equation variables
Equation sample 2000 2021
F-statistic 0.4072 Prob. F (6,10) 0.8582
Log likelihood ratio 4.8091 Prob. Chi-Square(6) 0.5685
Wald-statistic 2.4433 Prob. Chi-Square(6) 0.8748
Source: Eviews Software Package (Version 7)
The results of the CUSUM test buttresses that data was stable over time because the
cumulative sum line did not go outside the two critical lines as shown in figure 4.1 above.
This is also supported by results from the chow breakpoint test results which have an F-
statistic probability greater than five percent level of significance revealing that
coefficients are stable across all regimes. The null hypothesis that data is unstable over
time is therefore rejected in favour of the alternative hypothesis, concluding that data is
stable over time despite that fact that it passes through a period of high instability; that is
2007 and 2008.

4.4.0 Analysis of the effect of taxation and other macro-economic fundamentals on

Economic Growth regression results
The model was estimated using pooled Ordinary Least Squares and the results are as
illustrated in Table 4.10 below. Six models were run progressively and it was noted that
the R-squared was increasing from model 4.1 (0.5197) to 4.6 (0.9258). Such a scenario
shows that as the number of variables increases, their combined effect on economic
growth also increased over time. It is also crucial to note that the effect of taxation alone
on economic growth is around seventy-five percent, indicating that the current levels of
economic growth prevailing in the country are probably a reflection of the effect of tax
revenue. The results are as summarized below:

61
Table 4.10 Effect of taxation and other macro-economic fundamentals on economic
growth in Zimbabwe
Model 4.1
Variable(s) Coefficient Std. Error t-statistic Prob
INCT
34.2547 2.1707 15.7802 0.0000
2
R 0.5197
Adjusted R2 0.5197

Model 4.2
Variable(s) Coefficient Std. Error t-statistic Prob
INCT
27.3086 3.1280 8.7305 0.0000
VAT
1.5807 5.6908 2.7853 0.0114
2
R 0.6539
Adjusted R2 0.6366

Model 4.3
Variable(s) Coefficient Std. Error t-statistic Prob
INCT
22.7262 3.1744 7.1591 0.0000
VAT
3.6407 8.8908 4.0870 0.0006
CGT
-3.0608 1.1008 -2.7689 0.0122
2
R 0.7534
Adjusted R2 0.7275

Model 4.4
Variable(s) Coefficient Std. Error t-statistic Prob
INCT
22.7793 3.1437 7.2459 0.0000
VAT
3.6007 8.8108 4.0895 0.0007
CGT
-2.2008 1.3108 -1.6720 0.1118

62
INFL
-0.6330 0.5395 -1.1734 0.2559
2
R 0.7709
Adjusted R2 0.7328

Model 4.5
Variable(s) Coefficient Std. Error t-statistic Prob
INCT
63.7809 7.2388 8.8109 0.0000
VAT
3.0307 5.3108 5.7100 0.0000
CGT
-2.3809 8.4709 -0.2810 0.7821
INFL
-0.4391 0.3211 -1.3673 0.1893
INTR
-28.3740 4.8410 -5.8612 0.0000
2
R 0.9242
Adjusted R2 0.9063

Model 4.6
Variable(s) Coefficient Std. Error t-statistic Prob
C
16.4861 329.1420 -0.4937 0.6287
INCT
67.2955 10.4701 6.4274 0.0000*
VAT
2.8007 7.8208 3.5740 0.0028*
CGT
-1.6709 9.0309 -0.1850 0.8557
INFL
-0.4518 0.3478 -1.2988 0.2136
INTR
-26.3269 6.8570 -3.8394 0.0016*
FDI
13.4581 90.9513 0.1480 0.8843
2
R 0.9258
Adjusted R2 0.8961
F-statistic 31.1802
Prob (F-statistic) 0.0000
Schwarz Criterion 13.2368

63
 *Significant @ 1%, **Significant @ 5%, *** Significant at 5%
Source: Eviews Software Package (Version 7)
Using the statistics in the above table (Model 4.6) to fit in the model specified in chapter
three, the resulting model becomes as follows:
Economic Growth = 16. 4841 + 67.2955 INCT + 2.8007 VAT – 26.3269 INTR

4.4.1 Interpretation of R-Squared

An R-squared coefficient of 0.9258 obtained from the estimated model means that 92.58
percent of the regressors used in the model were able to explain changes taking place in
the dependent variable. More precisely, it implies that 92.58 percent of changes in
economic growth in Zimbabwe are being explained by changes in tax revenue and
macro-economic fundamentals. The result makes sense because there are other factors
that influence economic growth in Zimbabwe that were not incorporated in the model
estimated and this accounts for approximately 7.42 percent.

4.4.2 Interpretation of Adjusted R-Squared

Adjusted R-Squared measures the proportion of the dependent variable that explains the
independent variables. Model estimates results shows that 89.61 percent of the dependent
variable was able to explain changes in the independent variables thereby making it a
good model.

4.4.3 Interpretation of the F-statistic

The main goal of the F-statistic is of testing the fitness of the model under estimation.
Generally, an F-statistic of greater than five is recommended by literature for the model
to be regarded as fit for estimation. This research obtained an F-statistic of 31.1802 which
is by far greater 5, making this model a good fit for estimation.

4.4.4 Interpretation of regression results

Economic Growth = 16. 4841 + 67.2955 INCT + 2.8007 VAT – 26.3269 INTR
The standard error of estimation for this regression model is 4.16 and this shows that
there are relatively other types of tax that forms part of tax revenue in Zimbabwe as well
as other macro-economic and institutional factors that influences the growth of the

64
 Zimbabwean economy which are not part of this model. A constant coefficient of 16.48
percent explains the growth of the Zimbabwean economy without taking into effect all
the other factors. In other words, it is the starting point. The regression coefficient of
income tax is 67.30 percent meaning that holding all other things constant, a one percent
increase in income tax revenue might result in a 67.30 percent increase in the growth of
the Zimbabwean economy. This means that the growth of the Zimbabwean economy is
very responsive to changes in income tax revenue. Moreso, a regression coefficient of
2.80 of VAT implies that, holding all other things constant, a one percent increase in tax
revenue through VAT results in a 2.80 percent increase in the growth of the Zimbabwean
economy. This then shows that the growth of the Zimbabwean economy is not very
responsive to variations in tax revenue through VAT. Moreso, a regression coefficient of
-26.33 of interest rates means that holding all other things constant, a one percent
increase in interest rates retards growth of the Zimbabwean economy by 26.33 percent.
Thus, the growth of the Zimbabwean economy is very responsive to changes in interest
rates.

4.4.5 Income Tax and Growth of the Zimbabwean Economy

According to findings of this study, the relationship between income tax and economic
growth is positive and this contradicts theory which provides that tax impacts negatively
on the growth of the economy. However, some empirical findings support this as
evidenced by some discussions made in the empirical literature review section. However,
according to the multiplier effect high tax rates are expected to increase economic growth
as tax payers reduce consumption in favour of savings. Thus, tax revenue from both
individuals and business corporations is positively stimulating the growth of the
Zimbabwean economy. The graph below summarizes the relationship between income
tax and economic growth in Zimbabwe for the time period under study:

65
 1,600

1,400

1,200

1,000

800

600

400

200

0
00 02 04 06 08 10 12 14 16 18 20

GDP INCT

Source: Eviews Software Package (Version 7)

Figure 4.2 Graph illustrating the relationship between income tax and economic
growth
4.4.6 The Relationship between Value Added Tax and Economic Growth
Results indicate that VAT positively influences growth of the Zimbabwean economy.
This again contradicts the theoretical proposition that taxation negatively affects
economic growth. It nonetheless supports several empirical findings across the globe as
already highlighted in the empirical literature review of this study. Value Added tax is
very difficult to evade as it is levied on goods and services though it is exempted on some
basic items. The fact that every citizen needs to buy food stuffs for consumption makes it
a very difficult type of tax to avoid or evade. This makes it a very good type of tax for
analyzing the economic growth patterns prevailing in the country compared to other types
of taxes like presumptive tax which can easily be evaded. The diagram below shows the
relationship between VAT and economic growth:

66
 4,000,000,000

3,500,000,000

3,000,000,000

2,500,000,000

2,000,000,000

1,500,000,000

1,000,000,000

500,000,000

0
00 02 04 06 08 10 12 14 16 18 20

GDP VAT

Source: Eviews Software Package (Version 7)

Figure 4.3 Graph Depicting the Relationship between VAT and Economic Growth
4.4.7 The Relationship between Interest Rates and Economic Growth
Regression results have indicated that the relationship between interest rates and
economic growth is negative. Empirical findings have provided three versions of this
relationship, an indication that these findings are abreast with previous studies. Generally,
high interest rates make borrowing expensive and as a result business entities suffer a
major blow by being unable to access capital from local banks for fear of losses. At the
same time, high interest rates encourage lending and in this scenario, bank deposits are
expected to increase. In the Long-run interest rates are expected to fall in a free floating
economy through the forces of demand and supply for loans. On the other hand, low
interest rates discourage lending by banks and other surplus economic agents. It
discourages saving, making loans available for lending by banks to dwindle. In the long-
run interest rates are expected to increase further reducing growth of the economy. The
graph below illustrates the relationship between interest rates and economic growth:

67
4,000,000,000

3,500,000,000

3,000,000,000

2,500,000,000

2,000,000,000

1,500,000,000

1,000,000,000

500,000,000

0
00 02 04 06 08 10 12 14 16 18 20

GDP INTR

Source: Eviews Software Package (Version 7)

Figure 4.4 Graph showing the Relationship between Interest Rate and Economic Growth
4.5 Summary
Summary statistics from the run regression model indicated that 92.58 percent of changes
in the growth of the Zimbabwean economy are explained by changes in income tax, value
added tax and interest rates. Income tax and VAT produced a positive beta coefficient
implying that the two components of a tax system have a positive relationship with
economic growth. Interest rate have a negative effect on economic growth while the rest
of the variables included for estimation in the regression model showed that they do not
affect economic growth trends in Zimbabwe in any way and of these are capital gains tax,
inflation and FDI. As such they have no significant influence on economic growth in
Zimbabwe. The remaining 7.42 percent represents other factors influencing economic
growth in Zimbabwe. They might be tax components, institutional factors or macro-
economic fundamentals which do not form part of this study but remain relevant in
influencing economic growth. Theorists support both a feed-forward and feed-backward
relationship between taxation and economic growth. The next chapter looks at research
findings, conclusions, recommendations and suggestions for future research.

68
 CHAPTER V
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
5.0 Introduction
The previous chapter looked at data presentation, analysis and discussion. As such, this
section provides a thorough analysis on the effect of taxation on economic growth in
Zimbabwe for the time period 2000 to 2021 has been explored in this research. The final
chapter focused on summarizing key findings, conclusions inferred from research
findings, recommendations, suggestions for future research and summary of the chapter.
Study findings showed that growth of the Zimbabwean economy is influenced by income
tax, value added tax and interest rates prevailing in the financial markets. In order to
stimulate economic growth, the Zimbabwean authorities may need to increase tax
revenue from income tax and value added tax while at the same time injecting funds into
financial markets to ensure interest rates are kept at a minimum.

5.1 Summary of the study

In chapter one the researcher introduced the topic under study giving a detailed
background of the subject matter under study as well as presenting the research problem
which is the main goal for conducting the research. As part of the background to the
study, the effect of taxation on economic growth was explored at international level,
continental level and national level analyzing works done by different scholars on the
subject matter. Prominent scholars like McBride (2012) who asserts that taxation had
always had a negative effect on economic growth were also discussed and it was seen
that the effect of taxation on economic growth is multifold. Taxes are believed to distort
the behaviour of economic agents resulting in stunted economic growth. However, in
some instances it has been found out that taxation can positively influence the growth of
the economy through the multiplier effect.

In chapter two, the study looked at literature review on the effect of taxation on economic
growth. The chapter started by looking at the conceptual framework defining major terms
used in the study such as taxation, economic growth and tax revenue among others.
Theoretical literature review then followed and in this part, the main focus was on

69
explaining theories that assists in explaining the effect of taxation on economic growth.
These theories highlighted how taxation promotes or hampers economic growth in any
economy by providing how citizens views taxation as well as on how they respond to tax
laws levied by authorities in an economy. The chapter further looked at empirical
literature review, clearly looking at previously done studies on the subject matter which
further shed light on how taxation affects the growth of economies of different countries.
In this part, an analysis of both developed and developing countries was done with the
view of generating an informed comparative analysis. It was further seen that, in
developing countries taxation tends to have a negative effect on economic growth while
in developed countries, it has a positive effect. This could be due to the fact that,
developed countries have a well-established tax administration and infrastructure
resulting in high volumes of tax revenue which is then directed towards economic
growth. Moreso, in developed countries, the level of tax compliance is high than in
developing countries resulting in high tax revenue.

Chapter three dealt with research methodology and in this section, data collection tools
used in the study was explained including the reason for their choice. This study
employed quantitative research methods and as such used the positivism research
philosophy. The main reason for choosing this research philosophy was to avoid bias that
comes with qualitative research methods. The study also made use of triangulation in
selecting research designs. The descriptive, explanatory and case study research designs
were used in this study with the view of ensuring that the shortfall of one method is
overcome by the strength of another. An economic growth model for the Zimbabwean
economy was derived in this chapter as well as justification for the inclusion of each and
every variable in the model.

In chapter four, the results from diagnostic test results performed such as the
autocorrelation, multicollinearity test and stability tests among others were presented in
form of graphs and tables. It is in this chapter that regression results for all the models
were presented in a tabular form for easy analysis. The relationship between all
significant independent variables and dependent variables were also illustrated

70
 graphically for easy analysis in this section. The conclusions and recommendations in
this study are based on the findings in this section.

5.2 Summary of Key Findings on the effect of Taxation on Economic Growth

The major goal of the research was to investigate the effect of taxation on the growth of
the Zimbabwean economy for the time period 2000 – 2021 by way of econometric
models. The study also went on to have a look at other macro-economic fundamentals like
inflation and interest rate that might be responsible for the low levels of economic growth
prevailing in Zimbabwe. These were then tested quantitatively to see their effect on the
growth of the Zimbabwean economy. Research findings indicates that income tax, value
added tax and interest rate are the variables that might be influencing economic growth in
Zimbabwe. Moreso, it has been noted that income tax and value added tax positively
influence the growth of the Zimbabwean economy while interest rate is negatively
affecting it.

Results from the regression model also showed that about 92.58 percent of variations in
economic growth in Zimbabwe are accounted for through the causative effects of income
tax, value added tax and interest rates. Approximately sixteen percent of the prevailing
levels of economic growth in Zimbabwe have nothing to do with variables included in the
model. Income tax is responsible for current economic growth trends being witnessed in
Zimbabwe as it accounts for about sixty-seven percent according to research findings.
Apart from that, the study also proved that there are other factors that were not
incorporated into the model which are responsible for the current levels of economic
growth prevailing in Zimbabwe and this is to a tune of 7.42 percent only. It was also
discovered that inflation, capital gains tax and FDI have no effect on the growth of the
Zimbabwean economy as the three factors proved to have an insignificant influence on
GDP. Below is a summary of research findings as per each objective including
hypotheses test results:

Objective 1: To examine the extent to which income tax revenue affects economic
growth in Zimbabwe over the time period 2000 – 2021.

71
The study found out that income tax revenue has a strong positive significant influence on
the growth of the Zimbabwean economy as supported by regression results. Thus holding
all other things constant, a one percent increase in income tax revenue might result in a
67.30 percent increase in the growth of the Zimbabwean economy. This means that the
growth of the Zimbabwean economy is very responsive to changes in income tax revenue.
Thus the growth of the Zimbabwean economy is to a greater extent influenced by income
tax.

Objective 2: To critically evaluate the extent to which value added tax influences
economic growth in Zimbabwe over the time period 2000 – 2021.
The study also found out that value added tax has a weak positive significant influence on
the growth of the Zimbabwean economy. Thus holding all other things constant, a one
percent increase in value added tax results in a 2.80 percent increase in the growth of the
Zimbabwean economy. This then shows that the growth of the Zimbabwean economy is
not very responsive to variations in tax revenue through value added tax. This entails that
the growth of the Zimbabwean economy is slightly influenced by value added tax.

Objective 3: To assess the extent to which capital gains tax influences economic
growth in Zimbabwe over the time period 2000 – 2021.
The study also found out that capital gains tax has no significant influence on the growth
of the Zimbabwean economy. Thus the growth of the Zimbabwean economy is not being
influenced by changes in the level of capital gains tax.

Objective 4: To investigate the other factors that has an impact on economic growth
in Zimbabwe.
The study further found out that interest rate is the only other factor moderately
influencing the growth of the Zimbabwean economy as supported by a regression
coefficient of -26.33. Thus holding all other things constant, a one percent increase in
interest rates retards growth of the Zimbabwean economy by 26.33 percent. Thus, the
growth of the Zimbabwean economy is very responsive to changes in interest rates. The
study further proved that high interest rates are a huge threat to the growth of the
Zimbabwean economy stemming from their negative significant influence on economic

72
 growth. Such findings are consistent with available empirical evidence on the subject
matter.
5.3 Hypotheses Test Findings
Hypotheses test was carried out using restrictive coefficients for Wald tests at one
percent, five percent and ten percent level of significance. The main goal was to find out
the effect of taxation on economic growth on a preliminary basis. The following were the
findings per each test conducted:
Hypothesis 1

The first hypothesis sought to find out whether the three tax components which were
adopted for use in this study had a significant influence on economic growth. Hypothesis
test findings showed that the effect of taxation on economic growth is statistically
significant and this meant that taxation is expected to positively impact the growth of the
Zimbabwean economy as supported by a probability coefficient of 0.0000. The null
hypothesis that there is no significant relationship between taxation and economic growth
is therefore rejected in favour of the alternative hypothesis.

Hypothesis 2

The second hypothesis sought to investigate whether there was any significant
relationship between macro-economic factors and economic growth in Zimbabwe.
Hypothesis test findings showed that the effect of macro-economic factors on economic
growth is statistically significant and this therefore means that macro-economic factors
are expected to positively influence the growth of the Zimbabwean economy as supported
by a probability coefficient of 0.0000. The null hypothesis that there is no significant
relationship between macro-economic factors and taxation is therefore rejected in favour
of the alternative hypothesis.

Hypothesis 3

The third hypothesis sought to find out whether there was any significant relationship
between Foreign Direct Investment and economic growth in Zimbabwe. Hypothesis test
findings found out that the effect of FDI on economic growth is statistically significant as
supported by probability coefficient of 0.0000 and this therefore means that FDI is

73
 expected to have a significant influence on the growth of the Zimbabwean economy. The
null hypothesis that FDI has no significant relationship with economic growth is therefore
rejected in favour of the alternative hypothesis.

5.4 Conclusions
Basing on the findings presented above, the investigator inferred the following
conclusions on the subject of the effect of taxation on economic growth in Zimbabwe:
5.4.1 The growth of the Zimbabwean economy is greatly influenced by income tax, value
added tax and interest rates. Thus it can be concluded that taxation has a positive effect
on economic growth in Zimbabwe and such findings are consistent with those of
Babatunde et al. (2017) but contrary to those of Aydin and Esen (2019) as well as Dladla
and Hlalefang (2018) who vouch for a negative effect. Such findings are also not in
tandem with theory.
5.4.2 Value added tax has a positive significant effect on economic growth in Zimbabwe. This
move abreast with the findings of Akwe (2017) who found out that value added tax has a
positive effect on the growth of the Kaduna state of Nigeria.
5.4.3 Income tax has a statistically significant positive effect on the growth of the Zimbabwean
economy. Such findings conform to several former studies undertaken in this area of
study but contradict theory which suggests that taxation negatively influences economic
growth. Such results are not in tandem with those of Maganya (2020) in his study of the
Tanzanian economy in which he found out that income tax has a negative effect on
economic growth.
5.4.4 Interest rates have a negative significant effect on economic growth. The study further
indicated that high interest rates hurt the growth of the Zimbabwean economy and this is
the current situation in the Zimbabwean economy. Such findings are new in this study
area.
5.4.5 Foreign Direct Investment, capital gains tax and inflation have no significant effect on the
growth of the Zimbabwean economy. Such findings on inflation are contrary to those of
Dehghan and Nonejad (2015) who concluded that inflation has a negative effect on
economic growth in their research of the Iranian economy. Moreso, the results on capital

74
 gains tax also contradicts those of Andrasic et al. (2018) who posited that capital gains
tax has a positive significant effect on economic growth in a study of OECD countries.

5.5 Recommendations
Given the above study conclusions, the following recommendations are therefore made:
5.5.1 The government must implement measures that broaden the tax base instead of increasing
taxes. The main goal should be that of lowering tax evasion with the view of increasing
tax revenue. The government should also ensure it implement measures that promote
business activities in order to widen the tax base of the nation.
5.5.2 The government should promote the inflow of FDI in the country by sticking to policy
consistency and drop politics of personality. This would increase tax revenue for the
country by widening the tax base which would be directed towards growth of the
economy at large.
5.5.3 There is need to ensure that the supply of credit loans is raised in banks through
promotion of an increase in bank deposits. Bank deposits can be increased by ensuring
bank confidence is restored and ensuring more people open bank accounts locally. There
is also need to lower inflation in order to lower interest rates. This is so because lenders
will demand higher interest rates as compensation for the decrease in purchasing power
of the money they are paid in the future.
5.5.4 There is also need to ensure a good tax infrastructure is put in place so that maximum tax
revenue can be collected which can then be directed towards need areas of the economy
with the view of sprucing up economic growth.
5.5.5 The government must aim to increase income tax and VAT as opposed to other tax types
with the view of bringing about a more impactful and positive increase in economic
growth through the multiplier effect.

5.6 Suggestions for further Research

The major focus of this study was to investigate the effect of taxation on economic
growth in Zimbabwe using pooled regression ordinary least squares. The study found out
that it is income tax, value added tax and interest rates influencing the growth of the
Zimbabwean economy. The study was limited to establishing causal relationships only
without taking long-run and short-run relationships which is crucial in establishing

75
direction of causation critical in indicating bi-directional relationships between taxation
and economic growth. This therefore calls for further research on the same topic with a
different research methodology so that both short-run and long-run relationships could be
established between taxation and economic growth in the Zimbabwean context.

76
References
Abata. M. A (2014). The Impact of Tax Revenue on Nigerian Economy (A case of Federal Board
of Inland Revenue). Journal of Policy and Development Studies, volume 9, number 1.

Abomaye-Nimenibo. W (2018). An Empirical Analysis of Tax Revenue and Economic Growth in

Nigeria from 1980 – 2015. Global Journal of Human – Social Sciences: Political Sciences.

Adaramola. A. O and Dada. O (2020). Impact of inflation on economic growth: evidence from
Nigeria. Investment Management and Financial Innovations, 17(2), 1-13.
doi:10.21511/imfi.17(2).2020.01.

Acemoglu. D (2007). Introduction to Modern Economic Growth, 5th edition.

ActionAid International Briefing (2018). Capital Gains Tax, Progressive Taxation Briefing
Working Papers.

African Forum and Network on Debt and Development (AFRODAD), (2011). What has tax got
to do with development: A critical look at Zimbabwe‟s Tax System.

Ahmad. S, Sial. M. H and Ahmad. N (2018). Indirect Taxes and Economic Growth: An
Empirical Analysis of Pakistan. Journal of Applied Economics, volume 28, number 1 (65-81).

Alfaro, L. (2003). Foreign Direct Investment and Growth: Does the Sector Matter? (p. 32).
Harvard Business School.

Alinaghi, N. (2015). Taxes and Economic Growth in OECD Countries: A Meta-Regression

Analysis. NZAE (New Zealand Association of Economics) conference 2015, Wellington, New
Zealand.

Andrasic. J, Kalas. B, Mirovic. V, Milenkovic. N and Pjanic. M (2018). Econometric Modeling

of tax impact on economic growth: Panel evidence from OECD countries. Journal of Economic
Computation and Economic Cybernertics studies and Research, issue 4/2018, volume 52.

Anochiwa, L. I., & Maduka, A. (2015). Inflation and economic growth in Nigeria: empirical
evidence? Journal of Economics and Sustainable Development, 6(20), 113-121.

Anyaduba, J.O., 2004. Partnership taxation in Nigeria. ICAN Student Journal, 9(2): 15 - 17.

77
Atkinson, A. B. (1977). Optimal taxation and the direct versus indirect tax controversy,
Canadian Journal of Economics, 3(10), 590-606.

Aydin. C and Esen. O (2019). Optimal Tax Revenues and Economic Growth in Transition
Economies: A Threshold Regression Approach. Journal of Global Business and Economics
Review, volume 21, number 2.

Azubike, J.U.B., (2009). Challenges of Tax Authorities, Tax Payers in the management of Tax
Reform Processes. Niger Account 42 (2): 36-42.

Appah, E., & Oyandonghan, J.K. (2011). The Challenges of Tax Mobilization and Management
in the Nigerian Economy. Journal of Business Administration and Management, 6(2), 128-136.

Babatunde, O. A., Ibukun, A. O. and Oyeyemi, O. G., (2017). Taxation Revenue and Economic
Growth in Africa. Journal of Accounting and Taxation, 9(2), pp. 11-22.

Bakari. S, Ahmadi. A and Tiba. S (2020). The Nexus among domestic investment, taxation and
economic growth in Germany: Cointegration and Vector Error Correction Model Analysis,
Tunisisa.

Barro, R.J. (1990). Government spending in a simple model of endogenous growth. Journal of
Political Economy, Vol. 98, No. 5, pp. 103-125.

Bean. M. A (2017). Determinants of Interest Rates, Financial Mathematics Study Note.

Bhartia, H. L. (2009). Public Finance. New Delhi: Vikas Publishing House PVT Ltd.

Bleaney, M., Gemmell N., Kneller R. (2001). Testing the endogenous growth model: public
expenditure, taxation and growth over the long run. Canadian Journal of Economics, Vol. 34,
No. 1, pp. 36-57.

Breusch, T. and Pagan, A. (1980). The LM Test and Its Applications to Model Specification in
Econometrics. Review of Economic Studies, 47, 239-254.

Brautigam, D. (2019). Building leviathan: Revenue, state capacity, and governance. Institute of
Development Studies.33(4), 16-24.

78
Brown, R. L., Durbin, J. and Evans, J. M., (1975). Techniques for Testing the Constancy of
Regression Relationships Over Time. Journal of the Royal Statistical Society: Series B
(Methodological), 37(2), pp. 149-163.

Canicio. D and Zachary. T (2014). Causal relationship between government tax revenue growth
and economic growth: A case of Zimbabwe. Journal of economics and sustainable development,
5(17) 9-22.

Chude, D. I. and Chude, N. P. (2015). Impact of company income taxation on the profitability of
companies in Nigeria: A study of Nigerian Breweries, European Journal of Accounting, Auditing
and Finance Research, 3(8), 1 – 11.

Dehghan, M. and Nonejad, M (2015). The Impact of Tax Rates on Economic Growth of Iran in
the Years 1981-2010. Journal of Finance and Accounting, 3(6), pp. 220-226.

Dickey, D. A. and Fuller W. A., (1979). Distribution of the Estimators for Autoregressive Time
Series with a Unit Root. Journal of the American Statistical Association, 74(366), pp. 427-431.

Dladla. K and Hlalefang. K (2018). The impact of taxation on economic growth in South Africa.
MPRA Paper number 86219, Department of Economics, Nelson Mandela University, South
Africa.

Drobyshevsky. S, Trunin. P, Bozhechkova. A and Sinelnikova-Muryleva. E (2017). The effect of

Interest Rates on Economic Growth, Moscow.

Edame, E., Okoi, W. (2014). "The Impact of Taxation on Investment and Economic Development
in Nigeria". Academic Journal of Interdisciplinary Studies, 3(4), pp. 209-218.

Elshani. A and Ahmeti. S (2017). The effect of Progressive tax on economic growth: Empirical
evidence from European OECD countries. Journal of Economic Perspectives, volume 11, issue
4, 18-25.

Enejoh, S. Y., & Tsauni, A. M. (2017). An analytical study of the impact of inflation on
economic growth in Nigeria (1970–2016). International Journal of Academic Research in
Accounting, Finance and Management Sciences, 7(4), 110-120.

79
Faure. A. P (2014). Interest Rates, what are Interest Rates? SSRN Electronic Journal, Rhodes
University, South Africa.

Fischer, S. (1993). The Role of Macroeconomic Factors in Growth. Journal of Monetary

Economics, Vol. 47, No. 5, pp. 485-512.

Gale, W., Krupkin, A., Rueben, K. (2015). "The Relationship between Taxes and Growth: New
Evidence". National Tax Journal, 68(4), pp. 919-942.

Gale, W. G., Samwick, A. A. (2016). Effects of Income Tax Changes on Economic Growth. The
Brookings Institution.

Geetanjali, J V R and Venugopal, P. R., (2017). Impact of Direct Taxes on GDP: A Study. IOSR
Journal of Business and Management (IOSR-JBM), pp. 21-27.

Gentry. W. M (1999). A Set of Normative Prescriptions for Tax Policy usually based on
maximizing Social Welfare for a given Revenue Requirement.

Goode, R. (1984), Government Finance in Developing Countries, The Brookings Institution,

Washington D.C.

Gujarati, D., (2004). Basic Econometrics. 4th ed. s.l.:Tata McGraw Hill.

Ihenyen, C. J. and Mieseigha, E. G. (2014). Taxation as an instrument of economic growth (The

Nigeria perspective), Information and Knowledge Management, 4(12), 49 – 53.

Illyas, M. and Siddiqi, M. W (2010). The Impact of Revenue Gap on Economic Growth: A Case
Study of Pakistan. International Journal of Human and Social Sciences, 5:11.

Institute of Chartered Accountants of Nigeria (ICAN, 2006). Tax Management and Fiscal Policy
in Nigerian 1st edition.

Kalaš, B., Mirović, V. and Andrašić, J., (2017). Estimating the Impact of Taxes on the Economic
Growth in the United States. Economic Themes, 55(4), pp. 481-499.

Kalas, B., Mirovic, V., & Milenkovich, N. (2018). The relationship between taxes and economic
growth: Evidence from Serbia and Croatia. European Journal of Applied Economics, 15(2), 17–
28.

80
Kasidi, F., & Mwakanemela, K. (2015). Impact of inflation on economic growth: A case study of
Tanzania. Asian Journal of Empirical Research, 3(4), 363-380.

Kesner-Škreb, M. (1999). Porezna politika i gospodarski rast. Privredna kretanja i ekonomska

politika, Vol. 9, No. 73, pp. 62-121.

King, R., and S. Rebelo. (1990). „Public Policy and Economic Growth: Developing Neoclassical
Implications.‟ Journal of Political Economy 98:126–50.

Korkmaz. S, Yilgor. M and Aksoy. F (2019). The impact of direct and indirect taxes on the
growth of the Turkish economy. Journal of Public Sector Economics 43 (3) 311-323.

Kothari, C. R. (2004), “Research Methodology Methods & Techniques”, Second Edition, New
Delhi: New Age International Publisher.

Koutsoyiannis, A. (1977). “Theory of Econometrics: An Introductory Exposition of Econometric

Methods”.Hong Kong: MacMillan.

Krugman, P. (2012). More Laffering. The New York Times.

Kulu. E, Mensah. S and Senah. P. M (2021). Effects of Foreign Direct Investment on Economic
Growth in Ghana: The Role of Institutions, Journal of Economics Development, 20(1) pp.23-34.

Kunofiwa. T (2021). Tax Revenue and Economic Growth in Emerging Markets: Is Financial
Development Relevant? Journal of Accounting and Management, volume 11, number 1.

Iyke, B.N., &Takumah, W. (2015). The links between economic growth and tax revenue in
Ghana: an empirical investigation. University of South Africa, University of Cape Coast. Mpra
Paper No. 67281.

Maček, R. (2014). The Impact of Taxation on Economic Growth: Case Study of OECD
Countries. Review of economic perspectives, Vol. 14, No. 4, pp. 309-328.

Madalla, G. S. (1992). “Introduction to Econometrics. New York”: MacMillan Press.

Magamu. M. G (2010). The Relationship between government Revenue and Economic Growth in
Kenya. Dissertation submitted to University of Nairobi, Department of Finance and Accounting.

81
Maganya. M. H (2020). Tax Revenue and Economic Growth in a developing country: An
Autoregressive Distribution Lags Approach. Central European Economic Journal, 7(54), pp.
205-217, Tanzania.

Mamo, F. T. (2012). Economic growth and inflation: A panel data analysis (Published Master
Thesis Submitted Södertörns University, Department of Social Sciences, Economics).

McBride, W. (2012). What is the Evidence on Taxes and Growth? Tax Foundation, Special
Report 207, December 18, 2012.

Melnyk, L., Kubatko, O., & Pysarenko, S. (2014). The impact of foreign direct investment on
economic growth: case of post communism transition economies. Problems and perspectives in
Management, 12(1), 17–24.

Milenkovic. I and Kalas. B (2017). The impact of taxes on economic growth in OCED countries,
Barcelona. Journal of International Conferences on social issues and economic studies.

Muhammad DS, Lakhan R, Ghulam M, Numan (2013) Rate of interest and its impact on
investment to the extent of Pakistan. Pakistan Journal of Commerce and Social Sciences 7(1):91–
99.

Musgrave R. A. and Musgrave P. B. (1989). “Public Finance: In Theory and Practice”. Mc-
Graw-Hill International,NewYork, NY, USA.

Mushtaq V and Siddiqui. K (2016). Effect of Interest Rate on Economic Performance: Evidence
from Islamic and Non-Islamic economies. Journal of Financial Innovation, Pakistan.

Myles, G. D. (2007). “Economic Growth and the Role of Taxation”, University of Exerter and
Institute of Fiscal Studies Working Papers.

Nimani. A (2015). Taxes and their effects on Business Environment. Academic Journal of
Interdisciplinary studies, University of Gjakova, Italy.

Nnoruga. O. M, Chikwelu. M. M and Ndubuisi. A. N (2018). Effect of Tax Revenue on Economic

Development in Nigeria. International Journal of Research in Business, Economics and
Management, volume 2, issue 4.

82
Oboh. J. O, Chinonyelum. O. J, Edeme. R. K (2018). Tax Revenue and Economic Growth in
selected ECOWAS countries, Evidence from SURE Model. International Journal of Academic
Research in Accounting, Finance and Management Sciences, 8(3): pp. 310-324.

OECD (2008). Taxation and Economic Growth, Economics Department Working Papers, No.
620, OECD, Paris.

Ojo, S. (2003). Fundamental Principle of Nigeria taxation, Lagos, Sagribra Tax Publication.

Okofor, R. G (2012). Tax Revenue generation and Nigerian Economic development. European
Journal of Business and Management, Volume 4, number 19, pp. 49-56

Okwu, A., Oseni, I., & Obiakor, R. (2020). Does Foreign Direct Investment Enhance Economic
Growth? Evidence from 30 Leading Global Economies. Global Journal of Emerging Market
Economies, 12(2), 217–230.

Ogbonna, G.N. and Appah, E. (2012). Impact of Tax Reforms and Economic Growth of Nigeria:
A Time Series Analysis. Current Research Journal of Social Sciences, Vol.4(1):62-68.

Ouma, D. (2019). Revenue effects of tax reforms, economic growth and political environment in
Kenya, African Journal of Economic Review, 7 (1), pp. 136-153.

Omotoso, M. O. (2001). Principles of taxation. (1st ed.) Ibadan: D, First Shepherd Investment.

Palić, I., Ţmuk, B. and Grofelnik, B., 2017. The Long-Run Impact of Personal Income Taxation
on Economic Development: Evidence from Croatia. Croatian Review of Economic, Business and
Social Statistics, 3(1), pp. 35-44.

Pattanaik, Behera, Rajesh (2013). Real interest rate impact on investment and growth-What the
empirical evidence for India suggests. Reserve Bank of India.

Phiri, A., (2016). The Growth Trade-off between Direct and Indirect Taxes in 323 South Africa:
Evidence from a STR Model. Managing Global Transitions, 14(3), pp. 233-250.

Piana, V., (2003). Tax revenue. Economics Web Institute. Available from
http://www.economicswebinstitute.org/glossary/tax [Accessed 23th October, 2015].

Poulson W. & Kaplan G. (2008). State income taxes and economic growth. Cato journal, 28(1),
pp. 53-71.
83
Ramsey, F., A (1927). Contribution to the Theory of Taxation, Economic Journal, 37.

Riba. L (2016). The relationship between tax and economic growth: A South African
Perspective. Dissertation submitted to the Development Finance Centre graduate school of
Business, University of Cape town South Africa.

Salahuddin M, Islam MR, Salim SA (2009). Determinants of investment in muslim Developing

countries: An empirical investigation”. International Journal of Economics and Management
3(1):100–129.

Saunders, M. L. P and Thornhill. (2012). “A Research Methods for Business Students”,6th

Edition.

Shuaib, I. M., Augustine, O. E., & Frank, A. O. (2015). Impact of inflation rate on the economic
growth in Nigeria. British Journal of Economics, Management & Trade, 9(3), 1-11. https://doi.

org/10.9734/BJEMT/2015/15293.

Sion, B. (2019). DVLA sets pace for public sector, Wales Online, Retrieved 21 March, 2019

Smyth, D. (1994): "Inflation and Growth". Journal of Macroeconomics, Vol. 16, pp. 261-270,
spring.

Tobin J. (1965): Money and Economic Growth. Econometrica. Journal of the Econometric
Society, Vol. 33, No. 4, pp. 671-684.

Tokuoka, Kiichi (2012). Does the Business Environment Affect Corporate Investment in India?
IMF Working Paper, WP/12/70.

Tosun, M.S., & Abizadeh, S. (2005). Economic growth and tax components: An analysis of Tax
changes in OECD. Applied Economics, 37(19), 2251-2263. DOI:10.1080/00036840500293813.

United States of America Congressional Research Services (2021). Introduction to U.S

Economy: Inflation.

UNCTAD. (2021). “Foreign Direct Investment and Development”, UNCTAD Series on Issues in
International Investment Agreements. Geneva: UNCTAD.

Urio. F (2020). Effect of Mobile Payment System in Tax Collection in Tanzania.

84
Utz. S (2010). Ability to Pay Theory, Faculty Articles and Papers.

Von Soest, C., (2009). The African State and its Revenues: How Politics Influences Tax
Collection in Zambia and Botswana.

Weier. A (2006). Legal Definitions of Taxation Terms: Implications for the Design of
Environmental Taxes and Charges.

World Bank World Bank National Accounts data (various publications).

Yi, F., & Suyono, E. (2014). The relationship between tax revenue and economic growth of
Hebei province based on the tax multiplier effect, Global Economy and Finance Journal, 7(2): 1-
18.

Zimbabwe National Statistical Agency (ZIMSTATS), (2005-2017), Various issues. Quarterly

Digest of Statistics, Harare, Zimbabwe.

85
Appendices

Appendix 1: Unit Root Test Results

Gross Domestic Product

Null Hypothesis: D(GDP,2) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -6.867129 0.0000

Test critical values: 1% level -3.831511
5% level -3.029970
10% level -2.655194

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20
observations
and may not be accurate for a sample size of 19

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(GDP,3)
Method: Least Squares
Date: 05/25/22 Time: 13:28
Sample (adjusted): 2003 2021
Included observations: 19 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(GDP(-1),2) -1.470772 0.214176 -6.867129 0.0000

C 7.445502 33.30925 0.223527 0.8258

R-squared 0.735027 Mean dependent var 10.00000

Adjusted R-squared 0.719440 S.D. dependent var 274.0955
S.E. of regression 145.1826 Akaike info criterion 12.89316
Sum squared resid 358325.7 Schwarz criterion 12.99258
Log likelihood -120.4850 Hannan-Quinn criter. 12.90999
F-statistic 47.15746 Durbin-Watson stat 2.136187
Prob(F-statistic) 0.000003

86
Income Tax

Null Hypothesis: D(INCT) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.654420 0.0002

Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INCT,2)
Method: Least Squares
Date: 05/25/22 Time: 13:29
Sample (adjusted): 2002 2021
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INCT(-1)) -1.278074 0.226031 -5.654420 0.0000

C 0.551115 0.627144 0.878770 0.3911

R-squared 0.639802 Mean dependent var -0.063000

Adjusted R-squared 0.619791 S.D. dependent var 4.479799
S.E. of regression 2.762294 Akaike info criterion 4.964640
Sum squared resid 137.3449 Schwarz criterion 5.064213
Log likelihood -47.64640 Hannan-Quinn criter. 4.984078
F-statistic 31.97246 Durbin-Watson stat 1.879464
Prob(F-statistic) 0.000023

Value Added Tax

Null Hypothesis: D(VAT) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.564631 0.0020

Test critical values: 1% level -3.808546

87
 5% level -3.020686
10% level -2.650413

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(VAT,2)
Method: Least Squares
Date: 05/25/22 Time: 13:32
Sample (adjusted): 2002 2021
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(VAT(-1)) -1.071140 0.234661 -4.564631 0.0002

C 1.87E+08 83584827 2.243142 0.0377

R-squared 0.536511 Mean dependent var 1847638.

Adjusted R-squared 0.510761 S.D. dependent var 4.67E+08
S.E. of regression 3.27E+08 Akaike info criterion 42.14082
Sum squared resid 1.92E+18 Schwarz criterion 42.24039
Log likelihood -419.4082 Hannan-Quinn criter. 42.16025
F-statistic 20.83585 Durbin-Watson stat 1.973891
Prob(F-statistic) 0.000240

Capital Gains Tax

Null Hypothesis: D(CGT) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.372360 0.0030

Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(CGT,2)
Method: Least Squares
Date: 05/25/22 Time: 13:34
Sample (adjusted): 2002 2021

88
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(CGT(-1)) -1.030108 0.235595 -4.372360 0.0004

C 1.19E+09 1.12E+09 1.058063 0.3040

R-squared 0.515054 Mean dependent var -5571.850

Adjusted R-squared 0.488112 S.D. dependent var 6.82E+09
S.E. of regression 4.88E+09 Akaike info criterion 47.54971
Sum squared resid 4.29E+20 Schwarz criterion 47.64929
Log likelihood -473.4971 Hannan-Quinn criter. 47.56915
F-statistic 19.11753 Durbin-Watson stat 2.003530
Prob(F-statistic) 0.000367

Inflation
Null Hypothesis: D(INFL) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.490378 0.0003

Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFL,2)
Method: Least Squares
Date: 05/25/22 Time: 13:36
Sample (adjusted): 2002 2021
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INFL(-1)) -1.591550 0.289880 -5.490378 0.0000

C 22.83537 34.32006 0.665365 0.5143

R-squared 0.626124 Mean dependent var -21.14950

Adjusted R-squared 0.605353 S.D. dependent var 237.5701
S.E. of regression 149.2438 Akaike info criterion 12.94368
Sum squared resid 400927.0 Schwarz criterion 13.04325

89
Log likelihood -127.4368 Hannan-Quinn criter. 12.96312
F-statistic 30.14425 Durbin-Watson stat 1.992321
Prob(F-statistic) 0.000033

Interest Rate
Null Hypothesis: D(INTR) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.013282 0.0065

Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR,2)
Method: Least Squares
Date: 05/25/22 Time: 13:38
Sample (adjusted): 2002 2021
Included observations: 20 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(INTR(-1)) -0.943299 0.235044 -4.013282 0.0008

C 1.293577 1.113523 1.161698 0.2605

R-squared 0.472240 Mean dependent var -0.061000

Adjusted R-squared 0.442920 S.D. dependent var 6.358103
S.E. of regression 4.745546 Akaike info criterion 6.046930
Sum squared resid 405.3637 Schwarz criterion 6.146503
Log likelihood -58.46930 Hannan-Quinn criter. 6.066367
F-statistic 16.10644 Durbin-Watson stat 2.000605
Prob(F-statistic) 0.000815

90
Foreign Direct Investment
Null Hypothesis: FDI has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=4)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -4.471627 0.0022

Test critical values: 1% level -3.788030
5% level -3.012363
10% level -2.646119

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(FDI)
Method: Least Squares
Date: 05/25/22 Time: 13:39
Sample (adjusted): 2001 2021
Included observations: 21 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

FDI(-1) -1.025216 0.229271 -4.471627 0.0003

C 0.204543 0.094146 2.172608 0.0427

R-squared 0.512764 Mean dependent var 0.000476

Adjusted R-squared 0.487120 S.D. dependent var 0.526920
S.E. of regression 0.377357 Akaike info criterion 0.979144
Sum squared resid 2.705573 Schwarz criterion 1.078623
Log likelihood -8.281017 Hannan-Quinn criter. 1.000734
F-statistic 19.99544 Durbin-Watson stat 2.005702
Prob(F-statistic) 0.000261

91
Appendix 2: Heteroscedasticity Test Results
Heteroskedasticity Test: Breusch-Pagan-Godfrey

F-statistic 0.801471 Prob. F(6,15) 0.5838

Obs*R-squared 5.340761 Prob. Chi-Square(6) 0.5009
Scaled explained SS 3.180810 Prob. Chi-Square(6) 0.7858

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 05/25/22 Time: 13:58
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

C 95999.73 48396.11 1.983625 0.0659

INCT -2225.648 1539.491 -1.445704 0.1688
VAT 1.35E-05 1.15E-05 1.177473 0.2574
CGT -1.46E-06 1.33E-06 -1.097225 0.2898
INFL 35.91259 51.14479 0.702175 0.4933
INTR -779.9395 1008.227 -0.773575 0.4512
FDI -3618.580 13373.21 -0.270584 0.7904

R-squared 0.242762 Mean dependent var 12475.34

Adjusted R-squared -0.060133 S.D. dependent var 19161.92
S.E. of regression 19729.65 Akaike info criterion 22.87100
Sum squared resid 5.84E+09 Schwarz criterion 23.21815
Log likelihood -244.5810 Hannan-Quinn criter. 22.95278
F-statistic 0.801471 Durbin-Watson stat 1.539607
Prob(F-statistic) 0.583806

Appendix 3: Multicollinearity Test Results

a). Correlation Matrix

GDP INCT VAT CGT INFL INTR FDI

GDP 1.000000 0.895957 0.662592 0.286555 0.108266 0.346047 -0.149474
INCT 0.895957 1.000000 0.663403 0.322448 0.203174 0.524491 -0.026605
VAT 0.662592 0.663403 1.000000 0.821577 0.611892 0.822309 -0.218566
CGT 0.286555 0.322448 0.821577 1.000000 0.763221 0.767064 -0.178797
INFL 0.108266 0.203174 0.611892 0.763221 1.000000 0.602571 0.011738
INTR 0.346047 0.524491 0.822309 0.767064 0.602571 1.000000 -0.024331

92
 FDI -0.149474 -0.026605 -0.218566 -0.178797 0.011738 -0.024331 1.000000

b). Variance Inflation Factor (VIF)

Variance Inflation Factors
Date: 05/25/22 Time: 14:00
Sample: 2000 2021
Included observations: 22

Coefficient Uncentered Centered

Variable Variance VIF VIF

INCT 55.39227 57.03253 32.75670

VAT 3.09E-15 9.612965 6.65430
CGT 7.70E-17 7.318318 4.78643
INFL 0.114906 2.893730 1.89063
INTR 25.20359 57.63240 48.09801
FDI 7365.469 1.526630 1.09870

Appendix 4: Autocorrelation Test Results

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.607436 Prob. F(2,14) 0.5585

Obs*R-squared 1.754203 Prob. Chi-Square(2) 0.4160

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 05/25/22 Time: 13:54
Sample: 2000 2021
Included observations: 22
Presample missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

INCT 1.838831 7.813897 0.235328 0.8174

VAT 2.17E-08 6.04E-08 0.359817 0.7244
CGT -9.19E-10 9.04E-09 -0.101589 0.9205
INFL -0.100789 0.360591 -0.279511 0.7839
INTR -1.501000 5.326552 -0.281796 0.7822
FDI -22.23058 90.73396 -0.245008 0.8100
RESID(-1) -0.288053 0.295108 -0.976093 0.3456
RESID(-2) -0.258567 0.312773 -0.826694 0.4223

R-squared 0.079737 Mean dependent var -1.227474

93
Adjusted R-squared -0.380395 S.D. dependent var 114.3146
S.E. of regression 134.3085 Akaike info criterion 12.91344
Sum squared resid 252542.7 Schwarz criterion 13.31019
Log likelihood -134.0479 Hannan-Quinn criter. 13.00690
Durbin-Watson stat 2.124172

Appendix 5: Stability Test Results

a). Chow Break Point

Chow Breakpoint Test: 2008

Null Hypothesis: No breaks at specified breakpoints
Varying regressors: All equation variables
Equation Sample: 2000 2021

F-statistic 0.407210 Prob. F(6,10) 0.8582

Log likelihood ratio 4.809070 Prob. Chi-Square(6) 0.5685
Wald Statistic 2.443261 Prob. Chi-Square(6) 0.8748

b). CUSUM Test Results

12

-4

-8

-12
06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21

CUSUM 5% Significance

94
Appendix 6: Regression Summary Equations and Durbin Watson Test Results
Model 4.1
Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:45
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

INCT 34.25466 2.170735 15.78021 0.0000

R-squared 0.519658 Mean dependent var 925.2273

Adjusted R-squared 0.519658 S.D. dependent var 416.2419
S.E. of regression 288.4836 Akaike info criterion 14.21154
Sum squared resid 1747678. Schwarz criterion 14.26114
Log likelihood -155.3270 Hannan-Quinn criter. 14.22323
Durbin-Watson stat 0.247356

Model 4.2
Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:46
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

INCT 27.30858 3.127966 8.730459 0.0000

VAT 1.5807 5.6908 2.785286 0.0114

R-squared 0.653905 Mean dependent var 925.2273

Adjusted R-squared 0.636600 S.D. dependent var 416.2419
S.E. of regression 250.9216 Akaike info criterion 13.97467
Sum squared resid 1259233. Schwarz criterion 14.07385
Log likelihood -151.7213 Hannan-Quinn criter. 13.99803
Durbin-Watson stat 0.368762

Model 4.3
Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:47

95
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

INCT 22.72621 3.174430 7.159145 0.0000

VAT 3.6407 8.8908 4.086983 0.0006
CGT -3.0608 1.1008 -2.768878 0.0122

R-squared 0.753408 Mean dependent var 925.2273

Adjusted R-squared 0.727450 S.D. dependent var 416.2419
S.E. of regression 217.3044 Akaike info criterion 13.72660
Sum squared resid 897203.1 Schwarz criterion 13.87538
Log likelihood -147.9926 Hannan-Quinn criter. 13.76165
Durbin-Watson stat 1.022391

Model 4.4
Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:48
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

INCT 22.77933 3.143741 7.245932 0.0000

VAT 3.6007 8.8108 4.089522 0.0007
CGT -2.2008 1.3108 -1.672035 0.1118
INFL -0.633043 0.539510 -1.173367 0.2559

R-squared 0.770929 Mean dependent var 925.2273

Adjusted R-squared 0.732750 S.D. dependent var 416.2419
S.E. of regression 215.1813 Akaike info criterion 13.74380
Sum squared resid 833453.6 Schwarz criterion 13.94218
Log likelihood -147.1818 Hannan-Quinn criter. 13.79053
Durbin-Watson stat 0.820673

Model 4.5
Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:49
Sample: 2000 2021
Included observations: 22

96
 Variable Coefficient Std. Error t-Statistic Prob.

INCT 63.78094 7.238830 8.810946 0.0000

VAT 3.0307 5.3108 5.709962 0.0000
CGT -2.3809 8.4709 -0.280987 0.7821
INFL -0.439067 0.321122 -1.367291 0.1893
INTR -28.37402 4.841015 -5.861171 0.0000

R-squared 0.924168 Mean dependent var 925.2273

Adjusted R-squared 0.906326 S.D. dependent var 416.2419
S.E. of regression 127.3962 Akaike info criterion 12.72920
Sum squared resid 275906.4 Schwarz criterion 12.97716
Log likelihood -135.0212 Hannan-Quinn criter. 12.78761
Durbin-Watson stat 2.286652

Model 4.6 (a) Without Constant

Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:50
Sample: 2000 2021
Included observations: 22

Variable Coefficient Std. Error t-Statistic Prob.

INCT 63.75343 7.442598 8.566018 0.0000

VAT 3.06E-07 5.56E-08 5.507484 0.0000
CGT -2.06E-09 8.77E-09 -0.235184 0.8170
INFL -0.461424 0.338978 -1.361221 0.1923
INTR -28.56553 5.020317 -5.689985 0.0000
FDI 24.94298 85.82231 0.290635 0.7751

R-squared 0.924567 Mean dependent var 925.2273

Adjusted R-squared 0.900994 S.D. dependent var 416.2419
S.E. of regression 130.9717 Akaike info criterion 12.81484
Sum squared resid 274457.5 Schwarz criterion 13.11240
Log likelihood -134.9632 Hannan-Quinn criter. 12.88494
Durbin-Watson stat 2.315569

Model 4.6 (b) With Constant – Main Equation

Dependent Variable: GDP
Method: Least Squares
Date: 05/25/22 Time: 14:54
Sample: 2000 2021
Included observations: 22

97
 Variable Coefficient Std. Error t-Statistic Prob.

C 16.48614 329.1420 -0.493667 0.6287

INCT 67.29554 10.47008 6.427414 0.0000
VAT 2.80E-07 7.82E-08 3.574016 0.0028
CGT -1.67E-09 9.03E-09 -0.185022 0.8557
INFL -0.451764 0.347836 -1.298784 0.2136
INTR -26.32688 6.856956 -3.839442 0.0016
FDI 13.45807 90.95126 0.147970 0.8843

R-squared 0.925773 Mean dependent var 925.2273

Adjusted R-squared 0.896082 S.D. dependent var 416.2419
S.E. of regression 4.167814 Akaike info criterion 12.88963
Sum squared resid 270069.6 Schwarz criterion 13.23678
Log likelihood -134.7860 Hannan-Quinn criter. 12.97141
F-statistic 31.18024 Durbin-Watson stat 2.092491
Prob(F-statistic) 0.000000

Appendix 7: Wald Restrictive Coefficients for Hypotheses Tests

a). Hypothesis Test For Taxation

Wald Test:
Equation: Taxation

Test Statistic Value df Probability

F-statistic 1.2832 (3, 16) 0.0000

Chi-square 3.8532 3 0.0000

Null Hypothesis: C(1)=INCT, C(2)=VAT, C(3)=CGT

Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(1) – INCT 42.41343 7.442598

C(2) – VAT -1.4208 5.5608
C(3) – CGT -1.0708 8.7709

Restrictions are linear in coefficients.

Restrictions evaluated at observation (t = "2000")

98
b). Hypothesis Test for Macro-economic factors
Wald Test:
Equation: Macro-economic Factors

Test Statistic Value df Probability

F-statistic 1.7917 (2, 16) 0.0000

Chi-square 3.5717 2 0.0000

Null Hypothesis: C(1)=INFL, C(2)=INTR

Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(1) - INFL 59.27343 7.442598

C(2) - INTR -30.76000 5.56E-08

Restrictions are linear in coefficients.

Restrictions evaluated at observation (t = "2000")

c). Hypothesis Test Result for Foreign Direct Investment

Wald Test:
Equation: FDI

Test Statistic Value df Probability

t-statistic 8.551239 16 0.0000

F-statistic 73.12368 (1, 16) 0.0000
Chi-square 73.12368 1 0.0000

Null Hypothesis: C(1)=FDI

Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(1) - FDI 63.64343 7.442598

Restrictions are linear in coefficients.

Restrictions evaluated at observation (t = "2000")

99
 Appendix 8: Descriptive Statistics and Normality Test Results
GDP INCT VAT CGT INFL INTR FDI
Mean 9.252273 27.99045 1.1509 3.7409 47.90773 41.45864 0.195455
Median 1021.000 30.44000 5.2708 8.0108 3.255000 39.71500 0.110000
Maximum 1465.000 34.18000 3.7009 2.3210 557.2100 60.67000 1.720000
Minimum 357.0000 19.98000 1.4208 1.0708 -72.73000 30.76000 -0.020000
Std. Dev. 416.2419 4.500250 1.0809 7.9409 134.7826 8.196033 0.359560
Skewness -0.036864 -0.348102 1.134731 2.101927 2.747933 0.911567 3.671344
Kurtosis 1.312934 1.645369 3.311254 5.450911 10.53674 3.155728 16.15632

Jarque-Bera 2.613991 2.126415 4.810061 21.70607 79.75639 3.069062 208.0867

Probability 0.270632 0.345346 0.090263 0.000019 0.000000 0.215557 0.000000

Sum 20355.00 615.7900 2.5210 8.2310 1053.970 912.0900 4.300000

Sum Sq. 3638404. 425.2973 2.4419 1.3221 381493.4 1410.674 2.714945
Dev.

Observation 22 22 22 22 22 22 22

Normality Test Results

6
Series: Residuals
Sample 2000 2021
5 Observations 22

4 Mean -1.227474
Median -7.410395
Maximum 267.1588
3 Minimum -236.5545
Std. Dev. 114.3146
Skewness 0.428663
2
Kurtosis 3.270914

1 Jarque-Bera 0.741034
Probability 0.690377

0
-250 -200 -150 -100 -50 0 50 100 150 200 250 300

100
Appendix 9: Regression Output Graphs for Relationships between Variables and their
Gradients
GDP and Income Tax
1,600

1,400

1,200

1,000

800

600

400

200

0
00 02 04 06 08 10 12 14 16 18 20

GDP INCT

GDP and Value Added Tax

4,000,000,000

3,500,000,000

3,000,000,000

2,500,000,000

2,000,000,000

1,500,000,000

1,000,000,000

500,000,000

0
00 02 04 06 08 10 12 14 16 18 20

GDP VAT

101
GDP and Interest Rate
4,000,000,000

3,500,000,000

3,000,000,000

2,500,000,000

2,000,000,000

1,500,000,000

1,000,000,000

500,000,000

0
00 02 04 06 08 10 12 14 16 18 20

GDP INTR

Appendix 10: Raw Data

YEAR GDP INCT VAT CGT INFL INTR FDI

2000 563 21.34 142362429 106542774 4.48 30.76 0.11
2001 568 22.6 199661806 106654211 -37.2 33.17 0.06
2002 483 24.34 235294290 110045674 -34.45 36.48 0.05
2003 478 24.98 267470000 122432567 -8.57 37.73 0.03
2004 483 25.87 349664707 356700342 113.57 43.87 1.72
2005 477 24.87 380278770 387453234 -31.52 47.16 0.2
2006 448 26.87 389939946 234678654 32.97 48.2 0.01
2007 432 19.98 480364754 236876544 -72.73 32.87 0.06
2008 357 22.56 574208500 238967543 156.96 33.33 0.18
2009 772 21.87 288277183 206542774 6.22 34.12 0.01
2010 948 29.98 376763897 700682827 3.04 35.28 0.35
2011 1094 30.9 477896012 900744037 3.47 36.74 0.3
2012 1305 32.14 1250119582 1121995918 3.72 36.94 0.29
2013 1430 30.95 1321512400 1121995918 1.63 37.47 0.14
2014 1435 30.95 1433396800 1121995918 -0.21 41.7 0.37
2015 1445 34.18 1578422534 1296593904 -2.41 41.93 0.11
2016 1465 32.62 1652936750 1152611620 -1.56 42.33 0.14
2017 1336 32.62 1795066163 1243968814 0.91 43.9 0.01
2018 1352 33.32 2329980427 1882583181 10.61 48.43 0.04

102
2019 1156 30.95 2402347193 23213897711 255.29 49.53 0.02
2020 1128 30.95 3608670570 23213897611 557.21 59.48 -0.02
2021 1200 30.95 3702922709 23213897611 92.54 60.67 0.12
Source: World Bank, IMF and ZIMSTATS Quarterly Digests

103