TAX REFORMS & REVENUE MOBILISATION: A CASE STUDY …
Transcript of TAX REFORMS & REVENUE MOBILISATION: A CASE STUDY …
TAX REFORMS & REVENUE MOBILISATION: A CASE STUDY OF
THE MINING SECTOR OF GHANA
BY
HARRISON LOUIS APPIAH
(10224358)
THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA,
LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT
FOR THE AWARD OF MPHIL ECONOMICS DEGREE
JUNE, 2013
University of Ghana http://ugspace.ug.edu.gh
i
DECLARATION
I hereby declare that this thesis is the result of research undertaken by me towards the
award of the Master of Philosophy (MPHIL) degree in Economics in the Department of
Economics, University of Ghana under the supervision of the under signed lecturers.
…………………………..
LOUIS HARRISON APPIAH
(10224358)
SUPERVISORS
Name: DR. S.K.K. AKOENA
Signature: ……………………..
Date: ………………………….
Name: DR. D. K. TWEREFOU
Signature: .........………………………
Date: ……………….............................
University of Ghana http://ugspace.ug.edu.gh
ii
ABSTRACT
A strong and efficient tax system provides the basis for enhanced economic growth and
development. Ghana‘s fiscal structure prior to 1983 had generally been characterized by
low revenue. As a result, Ghana undertook a number of reforms prescribed by the
International Monetary Fund (IMF) and the World Bank under the Economic Recovery
Programme (ERP) and the Structural Adjustment Programme (SAP). Under the
programmes, numerous policies were amended to establish a more attractive investment
climate for foreign-owned mineral-exploration and extraction companies. The country‘s
mining industry has since expanded rapidly experiencing, by 2004, a fivefold increase in
annual gold output and big rises in bauxite, diamond, and manganese production.
However, at the same time, there is an on-going debate as to whether the country is
benefiting from mining operations in terms of tax revenue mobilization. The purpose of
this study is to assess the impact of fiscal regimes in the mining sector in Ghana on
revenue mobilization in the sector using tax elasticity and buoyancy ratios.
The Singer method of dummy variables was employed in order to make adjustment for the
effect of discretionary tax measures so as to compute and then compare buoyancy and
elasticity measures. The empirical results indicated that buoyancy estimates were higher
than elasticity estimates; and the short-run elasticities were lower than the static long-run
elasticities. Estimation results further showed that discretionary tax measures were
effective in mobilizing additional tax revenues and that the tax system was inelastic during
the period. The study recommends that there is the need to reduce capital consumption
University of Ghana http://ugspace.ug.edu.gh
iii
allowances, a holistic review of the principle of ring-fencing and identifying new sources
of taxation that are elastic.
University of Ghana http://ugspace.ug.edu.gh
iv
DEDICATION
I wish to dedicate this work to my parents, Mr. andMrs. Harrison Mensah and to my sister,
Richlove Harrison Mensah, for their sacrifices and support throughout my educational
career.
University of Ghana http://ugspace.ug.edu.gh
v
ACKNOWLEDGEMENT
I thank the Lord God Almighty for his grace, mercy, strength, kindness and love towards
me throughout my stay in school. Special thanks also go to my father, Mr. Edmund
Harrison Mensah for his financial support and helping me attain my educational dreams.
My gratitude also goes to my supervisors Dr.S.K.K AkoenaandDr.D.K Twerefou for their
support, encouragement and guidance. I am grateful to Mr. Yaw Benneh-Amponsah and
Abdallah Ali-Nakyeafor their useful comments and support while undertaking this
research. I extend my heartfelt gratitude to Mr. Kwame Aryeh of Merson Capital for his
encouragement and advice.ToMr. Abel Fumey of the Department of Economics,
University of Ghana for his time and knowledge imparted in this work.
Further, I wish to thank EsenamMedemeAgradi of the Minerals Commission of Ghana,
Mr. Emmanuel Danso of the Ghana Chamber of Mines for their respective roles in
sourcing for data for this work.
Finally, I wish to acknowledge the staff of Merson Capital who in one way or the other
contributed to the successful completion of this work and toNana YaaDarko for her
academic and emotional support.
However, though I received a lot of guidance and support from technocrats and colleagues,
I hereby declare total responsibility for any errors, omissions or misrepresentation that
may be found in this work.
Louis Harrison Appiah
June, 2013
University of Ghana http://ugspace.ug.edu.gh
vi
TABLE OF CONTENTS CONTENT PAGE
DECLARATION ............................................................................................................................. i
ABSTRACT .................................................................................................................................... ii
DEDICATION ............................................................................................................................... iv
ACKNOWLEDGEMENT .............................................................................................................. v
TABLE OF CONTENTS ............................................................................................................... vi
LIST OF TABLES ......................................................................................................................... xi
LIST OF FIGURES ...................................................................................................................... xii
ABBREVIATIONS ..................................................................................................................... xiii
CHAPTER ONE: INTRODUCTION ............................................................................................. 1
1.0 Background of the study ................................................................................................. 1
1.1 Research Problem ........................................................................................................... 4
1.2 Objectives of the Study ................................................................................................... 6
1.3 Justification and Significance of the Study ..................................................................... 7
1.4 Organization of the Study ............................................................................................... 8
CHAPTER TWO: FISCAL OVERVIEW .................................................................................... 10
2.0 Introduction ................................................................................................................... 10
2.1 Overview of the Mining Sector ..................................................................................... 11
2.1.1 Restructuring of Mining Sector Legislation ............................................................... 14
University of Ghana http://ugspace.ug.edu.gh
vii
2.2 Trends in Fiscal Regime ............................................................................................... 15
2.3 Taxes ............................................................................................................................. 17
2.3.1 Mineral Royalties ....................................................................................................... 17
2.3.2 Corporate Income Taxes. ........................................................................................... 18
2.3.3 Windfall or additional profits taxes ............................................................................ 19
2.3.4 Withholding taxes ...................................................................................................... 19
2.3.5 Capital Gain Taxes ..................................................................................................... 20
2.3.6 Employee Pay Roll Tax (PAYE) ............................................................................... 20
2.4 Mining Sector Fiscal Incentives.................................................................................... 21
2.5 Economic Impact of Fiscal Regimes ............................................................................ 23
2.5.1 Foreign Direct Investment.......................................................................................... 23
2.5.2 Revenue Performance ................................................................................................ 25
2.5.3 Output Trends............................................................................................................. 27
2.5.4 Export Earnings .......................................................................................................... 29
2.6 Measures Adopted to Enhance National Revenue Mobilization .................................. 30
2.7 Conclusion .................................................................................................................... 33
CHAPTER THREE: REVIEW OF LITERATURE ..................................................................... 34
3.0 Introduction ................................................................................................................... 34
3.1 Role of Taxation ........................................................................................................... 34
3.2 Theoretical aspects of tax reforms ................................................................................ 35
University of Ghana http://ugspace.ug.edu.gh
viii
3.2.1 Tax Reform and Optimal Taxation ............................................................................ 38
3.2.2 Measuring Tax Revenue Productivity ........................................................................ 43
3.3 Tax Types...................................................................................................................... 49
3.3.1 Royalties ..................................................................................................................... 49
3.3.2 Company Income Tax ................................................................................................ 51
3.3.3 Export and Import Taxes ............................................................................................ 52
3.4 Empirical Literature Review ......................................................................................... 52
3.4 Problems with tax reform.............................................................................................. 59
3.5 Lessons for Tax Reform................................................................................................ 60
3.6 Conclusion .................................................................................................................... 63
CHAPTER FOUR: METHODOLOGY& DATA USED IN THE STUDY ................................ 64
4.0 Introduction ................................................................................................................... 64
4.1 Analytical Framework .................................................................................................. 64
4.2 Tax Elasticity ................................................................................................................ 65
4.3 Buoyancy of Tax ........................................................................................................... 68
4.4 Model Specification ...................................................................................................... 70
4.5 The Dummy Variable Approach ................................................................................... 73
4.6 Estimation Techniques .................................................................................................. 74
4.6.1 Taxes Considered ....................................................................................................... 74
4.6.2 Choice of tax bases and Data Sources ........................................................................ 74
University of Ghana http://ugspace.ug.edu.gh
ix
4.7 Data Characteristics ...................................................................................................... 76
4.7.1 Unit Root Test for Stationarity ................................................................................... 76
4.7.2 Cointegration analysis ................................................................................................ 77
4.7.3 Error Correction Model (ECM) ................................................................................. 78
CHAPTER FIVE: PRESENTATION & DISCUSSION OF RESULTS ..................................... 79
5.0 Introduction ................................................................................................................... 79
5.1 Results of Unit Root Test .............................................................................................. 79
5.1.1 Results of Cointegration Test ..................................................................................... 81
5.2 Regression Results ........................................................................................................ 85
5.2.1 Tax Buoyancy: Long Run and Short Run .................................................................. 86
5.2.2 Decomposed Buoyancy into Tax-to-Base and Base-to-Income Elasticities .............. 89
5.2.3 Income elasticity of the Mining Tax System in Ghana .............................................. 93
5.3 Comparison of Tax Buoyancy and Elasticity ............................................................... 96
5.4 Conclusions ................................................................................................................... 96
CHAPTER SIX: SUMMARY, CONCLUSION AND RECOMMENDATIONS ....................... 99
6.0 Introduction ................................................................................................................... 99
6.1 Recommendations ....................................................................................................... 101
6.1.1 General Recommendations ...................................................................................... 101
6.1.2 Specific Recommendations ...................................................................................... 102
6.3 Limitations of the Study.............................................................................................. 105
University of Ghana http://ugspace.ug.edu.gh
x
REFERENCES ........................................................................................................................... 106
APPENDIX ................................................................................................................................. 112
University of Ghana http://ugspace.ug.edu.gh
xi
LIST OF TABLES
TABLE PAGE
Table 1.0: Tax revenue by Agencies (in % of total tax revenue) .................................................. 3
Table 2.0: Percentage shares of mining tax revenue to total GRA collections. ........................... 26
Table 2.1: Production of minerals ............................................................................................. 28
Table 2.2: Mineral and Total exports for selected years. (export value in million USD) ............ 29
Table 3.0: Summary of Some Key Empirical Studies ................................................................ 59
Table 4.0: Description of Various Taxes and Their Proxy Bases ............................................... 75
Table 5.0: Results of Augmented Dickey-Fuller (ADF) Test for Unit Root ............................... 80
Table 5.1: Engle-Granger Two Step Cointegration Test: Buoyancy Results .............................. 81
Table 5.2: Engle-Granger Two Step Cointegration Test: Tax-to-Base Elasticity Results ........... 83
Table 5.3: Engle-Granger Two Step Cointegration Test: Base-to-Income Elasticity Results ...... 85
Table 5.4: Estimated Tax Buoyancy Results of the Tax system in Ghana‘s Mining Sector ........ 87
Table 5.5: Estimated Tax-to-Base Elasticity of the Mining Sector Tax System ......................... 90
Table 5.6: Estimated Base-to-Income Elasticity of the Mining Sector Tax System .................... 92
Table 5.7: Estimated Tax Elasticity Results of the Tax system in Ghana‘s Mining Sector ......... 93
Table 5.8: Estimates of Long Run Tax Buoyancy and Elasticity ............................................... 96
University of Ghana http://ugspace.ug.edu.gh
xii
LIST OF FIGURES
FIGURE PAGE
Figure 2.0: Total Investment Inflow into the Mining Sector of Ghana ....................................... 24
Figure 5.0: Graphical Representation of the log of Tax variables and their respective bases. ..... 79
Figure 5.1: Co-integration Graphs for the Buoyancy Equation .................................................. 82
Figure 5.2: Cointegration Graphs for the Tax-to-Base Residuals ............................................... 84
University of Ghana http://ugspace.ug.edu.gh
xiii
ABBREVIATIONS
ADF Augmented Dickey Fuller
APT Additional Profit Tax
CEPS Customs, Excise and Preventive Service
DTMs Discretionary Tax Measures
DTRT Double Taxation Relief Treaties
ECM Error Correction Models
ERP Economic Recovery Programme
FDI Foreign Direct Investment
GoG Government of Ghana
GCNET Ghana Customs Management System
GDP Gross Domestic Product
GRA Ghana Revenue Authority
GSE Ghana Stock Exchange
GSS Ghana Statistical Service
HTSTD Historical Time Series Tax Data
IFC International Finance Corporation
IMF International Monetary Fund
IRS Internal Revenue Service
JCI Johannesburg Consolidated Investments
LTU Large Taxpayer Unit
MDGs Millennium Development Goals
MIGA Multilateral Investment Guarantee Agency
OLS Ordinary Least Squares
PAYE Pay-As-You-Earn
SIC Schwarz Information Criterion
TIN Tax Identification Number
VAT Value Added Tax
WTO World Trade Organization
University of Ghana http://ugspace.ug.edu.gh
1
CHAPTER ONE
INTRODUCTION
1.0 Background of the study
Ghana was among a host of developing countries that experienced fiscal crises in the
1970s and 1980s. These fiscal imbalances led to undesirable impacts on domestic prices,
interest rates and balance of payments. Thefiscal imbalances also led to differing policy
instruments for accelerating growth. There are a number of policy instruments that can be
used to accelerate growth. Tax reforms constitute a major type of policy instrument used in
Ghana. During the last decade, the country has consistently spent more revenue than it is
able to generate. These deficits have often been financed with foreign aid which underlines
the country‘s aid dependency. As the government needs to spend more on sectors such as
health, infrastructure and education, the critical issue has been how to effectively and
efficiently mobilize tax revenue using tax instruments that are least harmful to the poor.
This will thus involve reforming the tax system to ensure efficiency.
World Bank (1988) defines taxation as ‗compulsory, unrequited payments made to the
government by individuals, businesses or institutions‘. This underlines the principle that
taxation is compulsory and legal. There are four main components of taxes in Ghana,
namely, taxes on income and property, taxes on domestic goods and services, international
trade taxes and value-added tax. The value-added tax was the last one to be operational. It
was introduced initially in 1995, withdrawn and re-introduced in 1998. Prior to 1999 when
VAT became operational, taxes on international trade were the major component of tax
University of Ghana http://ugspace.ug.edu.gh
2
revenue in Ghana, followed by taxes on domestic goods and services and then taxes on
income and property. Total tax revenue for the year 2011 amounted to 17.5% of Gross
Domestic Product (GDP). Taxes on international trade as a proportion of total revenue
averaged 18.80% (Bank of Ghana, 2011).
Before 1990, macroeconomic analyses and projections were not exhaustively undertaken
to provide a base for effective and consistent fiscal policy formulation. The not too
successful fiscal performance in the period lies at the heart of the economic crisis the
country experienced. As the expenditure-revenue gap widened, deficit financing became
the principal source of the budgetary support, causing the share of government borrowing
from the domestic banking system mainly from the central bank, to increase from 49% in
1970 to 86% in 1982 (Kusi, 1991).
In 2001, the government opted for the Highly Indebted Poor countries initiative of the
World Bank and the IMF mainly because the economy was characterized by rapid
exchange rate depreciation, high inflation and very low external reserves. The severe crisis
was partly due to high debt service requirements that affected the budget of the
government and impacted negatively on the balance of payments (Bank of Ghana, 2003).
On revenue mobilisation, national tax collections have increased from about GHS 423.58
million in 2000 to about GHS 3971.42 million in 2008 at an average growth of about 33%.
In terms of collection of the various agencies as a percent of total tax, collection by
Customs Excise and Preventive Services (CEPS) alone is more than that of the other two
agencies. CEPS collections have averaged about 54% of total tax revenue. About 40% of
these collections come from Import VAT which is eventually credited to the accounts of
University of Ghana http://ugspace.ug.edu.gh
3
Agency 2000 2001 2002 2003 2004 2005 2006 2007 2008
IRS 31.9 32.4 33.0 32.0 32.0 31.3 31.0 30.0 33.0
VATS 13.6 11.6 12.7 11.9 12.7 13.6 15.0 15.3 15.5
CEPS 55.5 56.0 54.3 56.1 55.3 55.1 54.0 54.7 51.5
TOTAL* 423.58 655.70 860.81 1,278.28 1,686.50 2,057.60 2,370.82 3,040.23 3,971.42
Source: Twerefou et al, 2010 *Totals are stated in million GHS
VAT services in addition to the Domestic VAT thereby making VAT a better performing
tax in Ghana.
Table 1.0: Tax revenue by Agencies (in % of total tax revenue)
In spite of the efforts made, there exist some challenges in tax administration. Tax revenue
to GDP remains low at about 14% in 2007 compared to Sub Saharan African average of
18%. Also, the heavy reliance on Indirect Taxes poses a major challenge. About 70% of
total tax revenue in Ghana is comprised of indirect taxes - mainly VAT and trade taxes.
Revenue from direct taxes has risen over the last decade but has never exceeded 30% of
total tax revenue. Additionally, the narrow tax base presents a risk to the government and
poses a serious challenge. Disaggregated data reveals that the tax system depends on a
small number of large taxpayers who contribute the largest share of tax revenue – about
60% of the total income tax and 90% of the total turnover for VAT purposes.
The Mining sector of Ghana accounts for 27.61% of the country's total contributions to
GRA (Ghana Revenue Authority, 2011). Mining companies‘ tax obligations which cover
corporate tax, withholding tax and levies was 38.27% of the total company tax that GRA
collected in 2011. Ghana is Africa's 2nd largest gold producer, producing 80.5 t in 2008.
Ghana is also a major producer of bauxite, manganese and diamond. According to
University of Ghana http://ugspace.ug.edu.gh
4
Wikipedia.com, the country has over 23 large-scale mining companies producing gold,
diamonds, bauxite and manganese, and, there are also over 300 registered small scale
mining groups and 90 mine support service companies. Other mineral commodities
produced in the country are natural gas, petroleum, salt and silver. Ghana‘s economy
depends largely on exports of cocoa and gold.
The economic impact from the mining industry cannot be underestimated. For example in
2010, a 6.8% GDP growth was recorded by the economy on account on the performance
of the mining sector (GSS, 2011). An average of 25% was returned through the Bank of
Ghana to satisfy its statutory obligations and a further 43% through the commercial banks
with which the companies transact business. The inflows provided a significant cushion to
the local currency as it performed quite creditably against the country‘s trading currencies
(Chamber of Mines, 2010).
1.1 Research Problem
Wagner‘s law stipulates that public expenditure is a natural consequence of economic
growth (Demirbas, 1999). In an attempt to increase growth, successive governments have
raised public expenditure but have not been able to match it with revenue mobilization
through taxation and this has resulted in huge budget deficits. World Bank data of Ghana
indicates that tax revenue as a ratio of GDP is about 13.9% in 2008, lower than the Sub-
Saharan African average of 15.7%.
University of Ghana http://ugspace.ug.edu.gh
5
As Kaldor (1964) pointed out, the importance of public revenue to developing countries
can hardly be exaggerated if they are to achieve their hopes of accelerated economic
progress. It is important for a country like Ghana to reduce its dependence on external
finance as such financing may not be permanently available. The frequent deficit financing
and reliance on foreign aid call for increased efforts at generating revenues locally for
financing public expenditures. For example, between 2007 and 2010, Net Official
Development assistance and official aid received grew from 1.2 billion USD to 1.7 billion
USD in 2010 marking an increase of 417 basis points (World Bank Data).
World Bank data also indicates that Ghana‘s tax revenue as a ratio of total revenue is about
80% for 2011. Tax revenues from the mining sector as a percentage of total revenue raised
by Internal Revenue Service from 1990 to 2010 has averaged 12.62% (Minerals
Commission, 2010).
Various fiscal regimes have been put across by the tax authority as measures to boost
revenue generated from the mining sector. For a sector that has become synonymous with
Foreign Direct Investment, the focus has to be redirected towards ensuring that these
reforms attain the maximum possible beneficial impact through revenues and that the
growth of the mining sector is sustained in an orderly manner. It is therefore imperative
that a research be done to ascertain the buoyancy and elasticity of the tax system to
evaluate the revenue mobilisation capacity in the advent of these reforms.
The research questions are as follows:
What has been the process of tax reforms in the mining sector of Ghana?
University of Ghana http://ugspace.ug.edu.gh
6
What is the value of buoyancy and elasticity of individual mining taxes (specific to
the mining sector)?
Have these reforms improved revenue mobilisation?
What lessons do the outcomes have for policy makers in Ghana?
These are the research problems the study seeks to explore.
1.2 Objectives of the Study
In reducing the fiscal deficit as part of a structural adjustment programme, it is important
to be able to project what additional revenue can be mobilized within the existing tax
system as the economy grows. This requires an analysis of the revenues sources and their
responsiveness to GDP growth. Such an analysis will permit the identification of the
sources of fast revenue growth or, conversely, the causes of lagging revenue growth,
thereby suggesting measures to adopt to maximize revenue within the existing tax system
and/or the need to activate additional means of revenue generation.
The research;
Seeks to evaluate the tax revenue productivity of the mining sector tax system and
of individual taxes on the basis of estimates of tax elasticities and buoyancies. This
objective seeks to answer the research questions of the value of buoyancy and
elasticity of individual mining taxes.
University of Ghana http://ugspace.ug.edu.gh
7
Attempts to explain the links of tax reform measures to mining sector performance
and assess ways of mobilizing additional revenue. This objective will answer the
research question of whether the reforms have improved revenue mobilisation and
what lessons policy makers can draw.
1.3 Justification and Significance of the Study
Since the discovery of minerals in Ghana, the mining industry has played a tremendous
role towards the economic development of the nation. This is reflected in the sector‘s
contribution to foreign exchange, revenue, employment, GDP, and community
development through its corporate social responsibilities.
Mineral export for instance increased from 20% in the 1980s to 38% in 1998 of gross
foreign exchange earnings, with export earnings rising from USD 109.9 million in 1992 to
USD 717.8 million and further increased to USD 757 million in 2002, (Akabzaa, 2000).
Government revenues from royalties, PAYE, Levies and corporate taxes paid by mining
companies rose from GHS 2.5 million in 1994 to GHS 520 million in 2010. Also over the
same period, the contribution of royalty payments alone to total revenue collected by the
IRS rose from 4% in 1994 to 6% in 2010 (Ghana Chamber of Mines, 2010).
This study seeks to add to existing literature by bridging gaps in it. No study has been
conducted on the tax revenue productivity after the ERP on the individual sectors of the
Ghanaian economy.
University of Ghana http://ugspace.ug.edu.gh
8
In addition to bridging gaps in literature, this study will also enable us to better understand
the influence of the mining sector in terms of tax revenue performance.
In view of the role played by the mining sector in the economy, it is imperative that the
tax instruments are attractive enough to investors and at the same time generate a flow of
revenue for the government particularly because it is necessary that domestic resource
mobilisation is enhanced as a strategy for sustained growth and development. Indeed, as a
result of dwindling external resources, the Monterey Consensus (2002), Paris Declaration
(2005), Financing for Development in Doha (2008) and the Accra Agenda (2008)
underscored the need for efficient taxation systems to boost domestic revenue
mobilisation.
For the single act of policy direction, this research will identify which taxes are naturally
elastic or otherwise. With an inelastic tax system, a rising public expenditure may be
financed either through higher money supply with all the attendant problems of inflation
and balance of payment crises, or by annual upward revision of the existing tax rates. Also
an elastic tax system is efficient and acts as an instrument of stabilization.
1.4 Organization of the Study
The research is organized into six main chapters. Chapter 1 gives the motivation and
general background of the study. Chapter 2 looks at the fiscal overview of the Ghanaian
economy as well as the tax structure and reforms in Ghana with emphasis on the mining
sector. The third chapter makes an assessment of existing literature on tax reforms, both
University of Ghana http://ugspace.ug.edu.gh
9
theoretical and empirical. Chapter 4 and 5 consist of the research methodology adopted for
the study, focusing on the analytical framework, the empirical estimations and results
which entail model specification and estimation procedure respectively. The data source,
definition of variables and their proxies are also considered in the 4th
chapter. The sixth
summarizes and concludes the study with suggested recommendations and limitations of
the study.
University of Ghana http://ugspace.ug.edu.gh
10
CHAPTER TWO
FISCAL OVERVIEW
2.0 Introduction
Ghana faces daunting challenges of mobilising adequate resources to finance long-term
development, fight poverty and meet the main targets of the Millennium Development
Goals (MDGs) by the set deadline of 2015. With the deteriorating global financial system
that mitigated the chances of African countries to access external resources for
development goals, Ghana will have to focus on domestic revenue mobilisation as a better
alternative to meeting its development goals.
Taxes constitute a mainstay of domestic revenue yet Ghana which is endowed with
extractive resources which have attracted substantial Foreign Direct Investment has not
succeeded in mobilising substantial tax revenue from the extractive resources sector.
(Akabzaa, 2009).
The quantum of revenue that Ghana can exact from the mining sector will be dependent on
institutional and policy efficiency. Fiscal regimes define the quantum of taxes and tax
incentives for the mining sector.
University of Ghana http://ugspace.ug.edu.gh
11
2.1 Overview of the Mining Sector
Ghana like many African countries is endowed with enormous natural resources. Ghana is
the second largest gold producer in Africa after South Africa, the third biggest African
producer of manganese ore, and a significant producer of bauxite and diamonds.
Significant deposits of iron resources also exist. In addition, a number of industrial
minerals, including clays, dimension stone, limestone, common salt and silica sand, are
produced in commercial quantities. Ghana‘s enormous mineral endowment is credited to
the geological location. The country falls within the Precambrian rock shield of West
Africa that traditionally hosts most of the mineral resources of the continent.
According to Kesse (1985), when the Portuguese arrived in 1471, gold dust had been used
by the indigenous people for ceremonies and jewellery. Gold from Ghana accounted for 36
percent of total world gold output between 1493 and 1600 but declined to 3.70 percent in
2011(US Geological Survey, 2012). After independence, mining became one of the
government‘s main channels of mobilizing revenue and foreign exchange for its
developmental agenda. Many state owned exploration, mining and commodity buying
companies were established by the first post independence government of Ghana. This was
in consonance with government‘s belief of permanent sovereignty over the country‘s
natural resources.
In the early 1980s, as part of the structural adjustment programmes, the mining sector
received special attention with huge doses of financial and technical assistance from the
World bank, the International Monetary Fund, and other bilateral lenders, mainly
governments‘ of major transitional mining companies. The main objectives of the reforms
University of Ghana http://ugspace.ug.edu.gh
12
included; halting the decline in production by attracting international financial resources to
existing mines for rehabilitation of equipment and machinery as well as upgrading mine
infrastructure; improving management practices in state owned mines; instituting
macroeconomic and legal measures to attract investment in exploration for new mines and
expansion of old ones. The government‘s role was thus constrained to providing
administrative and regulatory oversight to the sector.
During the first years of the programme, the strategy sought to increase the worth of
existing mines through rehabilitation and to promote the opening up of new mines by
private companies, through government facilitated participation of the International
Finance Corporation (IFC). Ashanti Goldfields, which had the government of Ghana as
55% majority shareholder, enjoyed loans from multilateral and bilateral financial agencies
facilitated and guaranteed by the government. The company received a total of over $260
million, financed by a consortium of banks led by the IFC during this period for expansion
and rehabilitation of its operations from 1986 through to 1995. The expansion programme
included the introduction of open pit mining and the introduction of bulk cyanide heap
leaching operations and the provision of pollution abating technology including the
installation of bioxidation plant for the treatment of sulphide rich gold ore (Akabzaa et al,
2005). The rest of the existing mines, which were wholly state-owned such as the former
State Gold Mines at Tarwka, Dunkwa and Prestea); the Ghana Consolidated Diamond
Limited and Ghana Manganese Company Limited were given out to various groups of
investors under management contract agreements.
Having increased the worth and efficiency of the mines, the second phase of the reform
strategy involved the divesture of the state‘s interest in these mines. In all cases, but
University of Ghana http://ugspace.ug.edu.gh
13
Ashanti Goldfields Limited and Ghana Bauxite Company Limited, there was complete
divesture of hitherto state-owned mines to the private sector with government maintaining
a statutory 10% free carried equity as required by the Minerals and Mining Law.
Foreign multinational companies who had management contract agreements eventually
bought these mines, except in cases where they found them unviable. For instance,
Goldfields South Africa ran the Tarkwa mine on management contract from 1993 to 1994
and eventually purchased it in 1995. Johannesburg Consolidated Investments (JCI),
another South African company, ran the Prestea mine on contract from 1995 to 1996 and
purchased it in 1997. The Ghana National Manganese Corporation was taken over by a
group of Norwegian and German investors in 1993, who eventually acquired it in 1995.
However Ghana Consolidated Diamonds limited, which was managed by De Beers, the
South African diamond mining and trading giant, on contract, had failed to attract buyers
and De Beers refused to exercise its option to purchase it. It has since remained on the
divestiture list. The government also provided and continued to provide sovereign
guarantees with assistance from IFC, MIGA and other international credit agencies for the
smooth take-off some of the new mines such the Ghana Australia Goldfields Limited, now
AngloGold Ashanti (Iduapriem) Mine Limited.
In the case of Ashanti Goldfields Corporation, the government progressively reduced its
stake to 19% in 1998, from its original 55% through the sale of its shares initiated in 1993,
while in the case of Ghana Bauxite Company, the government reduced its shareholding
from 55% to 20% in 1998.
University of Ghana http://ugspace.ug.edu.gh
14
2.1.1 Restructuring of Mining Sector Legislation
In response to the cry of investors with respect to the regulation of the mining sector by
pieces of legislative and cumbersome procedures as well as concerns by the international
community in revitalizing the sector, a number of legislative instruments were
promulgated and passed from 1983 (Twerefou et al, 2007).
Notable among the legislative instruments was the PNDC Law 153 – Mining and Minerals
Law, 1986, PNDC Law 218 – Small Scale Mining Law, 1989 and Act 475 – Minerals and
Mining (Amendment) Act, 1994.
With respect to fiscal regimes, the following legislative instruments were promulgated; L.I
1349 – Mineral Royalty Regulation, 1987, Act 592 – Internal Revenue Act, PNDC Law
122 – Additional Profit Tax law, 1985.
A programme was put in place to develop the mining sector support institutions. This was
aimed primarily to;
Develop the capacity of the sector institutions to enable them better regulate
investment in an environmentally friendly manner.
Develop mechanisms to provide productivity and financial viability and
Institute measures to reduce the environmental impacts of small scale mining
operation (Minerals Commission, 1995).
To regulate the activities of the mining firms, other legislative instruments such as PNDCL
219 – Precious Minerals Marketing Corporation law, 1989, PNDCL 154 – Minerals
Commission Act, 1986, L.I 1652 – Environmental Assessment Regulation Act, 1994.
University of Ghana http://ugspace.ug.edu.gh
15
2.2 Trends in Fiscal Regime
The policy reforms in the mining sector included significant changes in the fiscal
provisions for investments in the mining sector, which have arguably been more generous
to multinational mining companies. Liberalization of the mining sector was to create an
enabling mining environment to attract foreign investment into the sector.
These included a progressive reduction in corporate income tax, from 50-55% in 1975 to
45% in 1986 and further reduced to 35% in 1994 and down to 22% and 25% for 32
companies listed on the Ghana stock exchange and those not listed, respectively. Mineral
royalties which stood at 6% of total value of minerals won in 1975 was reduced to between
3% -12% in 1987 and further down to between 3%- 6% with the enactment of Act 703, in
2006. Other duties such as the 5% mineral duty, 5-35% import duty, and 33- 75% foreign
exchange tax provided for in 1975 and maintained in the investment code of 1981 were
scrapped with the enactment of PNCL 153 and Act 703 of 1986 and 2006 respectively,
while an additional profit tax, designed to capture windfall profits in times high
commodity prices, such as is the case in the gold sector, which was introduced in 1986 was
scrapped in 2006.
In 1975 a decree provided for a modest depreciation regime of 20% in the first year of
investment and 15% for subsequent annual allowance to enable investors to recoup their
initial capital expenditure. The same quantum of depreciation allowance was maintained in
the 1981 code. However, it was increased to 75% for the first year and 50% for subsequent
years in PNDCL 153, 1986. This was again effectively increased to 80% for the first year
University of Ghana http://ugspace.ug.edu.gh
16
and 50% in subsequent years by the provisions of the Internal Revenue Act (Act 592) of
2000, which is referenced in the 2006 Minerals and Mining Act. In the same vein, mining
companies were exempted from payment of custom import duties on plant, machinery,
equipment, and accessories imported for use in the mining sector by the provisions of the
1986 and 2006 codes. Staff of mining companies was also exempted from paying income
taxes related to furnished accommodation at mine site. Personal remittance quota for
expatriate‘s personnel was free from tax imposed for the transfer of external currency out
of the country. Act 703 permits the holder of a mining lease to open and retain in an
offshore account, an amount not less than 25% of the foreign exchange for the acquisition
of spare parts, raw materials, and the machinery and equipment; debts servicing and
dividend payment; remittance in respect of quotas for expatriate personnel and the transfer
of capital in the event of a sale or liquidation of the mining operations. In the same vein,
government equity participation which stood at 55-100% in 1975 and 1981 was reduced to
10% in 1986 and same maintain in the 2006 mining act.
The table below summarises the Post and Pre-ERP Mining Sector Fiscal Regime
University of Ghana http://ugspace.ug.edu.gh
17
Table 2.0: Pre & Post ERP Mining Sector Fiscal Regime
Source: Akabzaa et al, (2009)
2.3 Taxes
2.3.1 Mineral Royalties
Mineral royalty is a production based tax which is levied on the basis of Article 25 of the
Minerals and Mining Act, 2006; ACT 703 which states that ‗A holder of a mining lease,
restricted mining lease or small scale mining licence shall pay royalty that may be
prescribed in respect of minerals obtained from its mining operations to the Republic,
except that the rate of royalty shall not be more than 6% or less than 3% of the total
revenue of minerals obtained by the holder.‖ Mining Companies are liable to pay royalties
Items SMCD 5 (1975) Act 437
Investment
Code 1981
PNDCL
153 1986
Amend-ments
to law 153
Act 703, 2006
Incentives
Initial capital allowance 20% 20% 75% 75%
Subsequent capital allowance 15% n/a 50% 50%
Investment allowance 5% n/a 5% 5%
Carried forward losses for purposes of n/a n/a Up to 5 Up to 5 years
Off-shore retention of sales n/a n/a 25%-80% 25%-80%
R&D allowance n/a 25% Exempt Exempt
Taxes
Mineral Duty 5-10% 5-10% Exempt Exempt
Import duty 5-35% 5-35% Exempt Exempt
Foreign exchange tax 33-75% 33-75% Exempt Exempt
Import license tax or import levy 10% 10% Exempt Exempt Exempt
Gold export levy $3/Oz for every
oz>100,000
$3/Oz for every
oz>100,000
Exempt Exempt Exempt
Corporate income tax 50-55% 45% 45% 35% 25%, 22%
Royalty 6% 2-6% 3-12% 3-6%
Withholding tax 15% 15% 10% 10%
Capital Gain tax 15% 15% 10% 10%
Additional Profit Tax (APT) n/a n/a 25% 0%
Minimum Turnover tax (National 2.50% 2.50% n/a 2% of before 0%
VAT n/a n/a n/a Exempt Exempt
NHIS n/a n/a n/a Exempt Exempt
Govt. Equity participation in mining lease 55-100% 55-100% 10% 10%
University of Ghana http://ugspace.ug.edu.gh
18
immediately they commence mineral production. The percentage (3%-6%) to be applied is
based on the profitability
(defined as the operating ratio, which is the ratio between the operating margin and the
gross value of minerals extracted) of the lease holder‘s operations. The operating margin is
the value of total minerals extracted less the operating cost. The lowest rate of 3% is
payable unless the operating ratio exceeds 30% when it increases progressively up to 6%.
2.3.2 Corporate Income Taxes.
The income tax rate has undergone several changes over the years. The rate used to be
45%. Aside the usual deductions, a company is also allowed to capitalize all expenditure
during reconnaissance, prospecting and development in accordance with the capital
allowance (amortization) provisions of the law. The corporate tax rate in Ghana has been
on a declining rate. The essence is also to attract investors into the country just like the
royalty rate. For instance, the rate which used to be 32.5% in 2001 has declined to 25% by
2006.
As of 2010, mining companies listed on the Ghana Stock Exchange pay on their
chargeable income at a rate of 22%, while those not listed pay 25%.
University of Ghana http://ugspace.ug.edu.gh
19
2.3.3 Windfall or additional profits taxes
Windfall taxes, otherwise known as Additional Profits Taxes (APT) are levies
governments employ to collect extra rent generated by mining companies during times of
windfall profits, arising from high mineral prices or discovery of usually high quality
deposits. The 2006 Mineral and Mining Act scrapped the additional profits tax provision in
the 1986 mining code. Many African countries removed the law from their status, because
it remained unapplied for a long time due to sustained period of depressed prices.
However from the year 2002 to 2008, these countries could have mobilized substantial
revenue, from APT if they had maintained the provisions in their tax status. During this
period mining companies across the globe made huge profits, above the average threshold
of 25%, typically required for the application of the tax. The period marked unexpected
steep increases in international prices of gold, copper, particularly, with mining companies
recording annual net return on investment of more than 35% (PriceWaterHouse Coopers,
2007).
2.3.4 Withholding taxes
Withholding taxes are levied on the services provided by non-taxpayers in a given
jurisdiction, paid directly by mining companies to tax authorities. The tax law provides for
withholding tax on dividends paid to shareholders, loan interest, and fees paid to
consultants of between 8% and 15%.
University of Ghana http://ugspace.ug.edu.gh
20
2.3.5 Capital Gain Taxes
These taxes are levied on mining plants and equipment which have been disposed of for
amounts above their stipulated book value. Apart from mining plant and equipment,
mineral properties that are sold above their stipulated book value should typically attract
capital gain tax. In Ghana, several mining properties have frequently been disposed
however, according to Boas and Associates (2007), apart from a recent transaction
involving Newmont Ghana Limited and Normandy Plc, for which capital gain taxes were
paid, no mining company paid capital gain taxes. The tax laws provide for capital gain tax
rate of 10% on all such transactions.
2.3.6 Employee Pay Roll Tax (PAYE)
This is tax on emoluments of employees and is not limited to mining. It is however a
significant source of revenue from the mining sector given the relatively higher incomes of
employees of the mining sector. It is largely derived from Ghanaian employees, as
expatriates have been exempted as part of the fiscal incentives to multinational mining
companies. The rate varies according to an individual‘s income level from 0% for those
receiving the minimum wage to a maximum of 25%.
University of Ghana http://ugspace.ug.edu.gh
21
2.4 Mining Sector Fiscal Incentives
To attract investment and to reduce the overall tax burden of an enterprise in the mining
sector, tax incentives are designed. They are particularly provided in the mining sector to
make exploration and production of minerals more attractive. According to Akabzaa et al
(2009), these incentives are provided with the purpose of attracting investment to remote
and deprived regions of the country; enhance the performance of the mining sector and
lead to overall maximisation of government revenue and other economic benefits such as
employment or to enhance technology transfer. Tax incentives to the mining sector is
defined by the Minerals and Mining Act, 2006 (Act 703) and the Internal Revenue Act,
2000 (Act 592). These incentives include:
Capital Allowance
Mining companies receive higher rates for capital allowances under the income tax code,
Internal Revenue Act, 2000 (Act 592). For assets like buildings, plant and machinery used
in mining operations, a rate of 80% of the cost base is granted as capital allowance in the
year of acquisition, 50% is further granted on the reducing balance basis in the subsequent
year. In addition, 5% of the cost base of the previous year is added to the written down
value of the immediate succeeding year before any capital allowance is granted for the
year. This means that the investor effectively recoups 105% of any investment in
qualifying mining expenditure, thus ensuring a very short pay-back period. Furthermore,
mining companies are permitted to carry forward unutilised accumulated capital
allowances forever until they are utilised.
University of Ghana http://ugspace.ug.edu.gh
22
Carry Forward Losses
This incentive allows losses arising in any tax year to be carried forward for deductions
against income in the subsequent year. It also provides that such losses be deducted within
five years following the year in which the loss is incurred and be ducted in order in which
it was incurred.
Custom Duty Exemptions
Mining companies are exempted from import duties on plant, equipment and accessories
exclusively for the purposes of mining operations. The list of items that are zero rated for
custom duty purposes are typically contained a mining list. Mining companies are however
required to pay exemption fees.
Personal Remittance Quotas for Expatriates
Expatriates are exempt from taxes imposed by an enactment regulating the transfer of
money out of the country. In other words, expatriates in the mining sector can transfer
funds abroad without paying taxes.
University of Ghana http://ugspace.ug.edu.gh
23
Double Taxation Reliefs
Large scale mining companies are usually foreign multinational firms operating through
their subsidiaries. These companies are required to pay taxes in Ghana and may be
required to pay in their home countries. This amounts to double taxation. Ghana‘s tax code
provides for relief of double taxation under which companies from countries that have
signed Double Taxation Relief Treaties (DTRT) with Ghana maybe granted this relief.
United Kingdom, South Africa, Italy, Belgium, France are among other countries that have
DTRT with Ghana.
Others include development or investment agreements, transferability of capital, enhanced
immigration quotas, tax free accommodation for mine employees.
2.5 Economic Impact of Fiscal Regimes
2.5.1 Foreign Direct Investment
Foreign Direct Investment (FDI) has been viewed as a major stimulus to economic growth
in developing countries. Its ability to deal with two major obstacles, namely, shortages of
financial resources, and technology and skills, has made it the centre of attention for policy
makers in low-income countries. Ghana has arisen as one of the most attractive destination
of investment largely due to the rich geological data and conducive investment
environment in the mining sector. For example, according to Tsikata (1997), the country
was ranked the third most attractive country for mining investment after Botswana and
Zimbabwe.
University of Ghana http://ugspace.ug.edu.gh
24
Source: Ghana Minerals Commission, 2011
0
100
200
300
400
500
600
700
800
900
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
in m
illio
n U
SD
Investment Inflow
The issue of FDI is becoming synonymous with the mining sector in Ghana. This is in
view of the large proportion of total private foreign capital that has been directed into the
sector over the years. Between the mid 1980s and 2005, investment into the sector totalled
well over USD 5 billion with an average inflow of well over USD 200 million. The non-
mining sector recorded accumulated net flows of a little over USD 1.6 billion between
1980 and 2003.
Abdulai (2005) attributed this phenomenon to the benefits the mining sector had received
from financing under the ERP and which had it turn been responding most positively in
terms of output and foreign exchange earnings. The introduction of the new mining
technologies and improved price on international commodity markets coupled with
profitability of the sector played a crucial role in attracting investors into the sector.
Figure 2.0: Total Investment Inflow into the Mining Sector of Ghana
University of Ghana http://ugspace.ug.edu.gh
25
Cumulatively, investment inflow into the mining sector from 2000 to 2010 stood at
approximately USD 6.4 billion.
2.5.2 Revenue Performance
The mining sector contributes to government revenue through the payment of direct taxes
such as corporate tax, taxes on wages and salaries of employees, royalties and dividends.
The share of mining in total collections from the Ghana Revenue Authority (GRA)
increased from 8.9% in 1990 to 21.3% in 2010. On the average, mining has accounted for
about 13.3% of the total GRA collection in 2000s.
The mining industry contributes to government revenue through the payment of direct
taxes such as corporate tax, taxes on wages and salaries of employees, royalties and
dividends. In Ghana, direct tax which is mainly administered by the Ghana Revenue
Authority forms between 25-30% of total tax revenue. Data on the contribution of the
mining sector to government revenue in the 1980s are made up of pieces of information,
quite scanty and unreliable at best. However, statistics from the GRA shows that the
mining sector contributed considerably to direct taxes in terms of corporate tax, income
tax, royalties and the introduction of the reconstruction levy. The share of mining in total
GRA collection increased from 8.9 percent in 1990 to 21.3 percent in 2010. On the
average, mining has accounted for about 13.3 percent of the total GRA collection in the
2000s.
University of Ghana http://ugspace.ug.edu.gh
26
YEAR
CORPORATE
TAX
MINERAL
ROYALTIES PAYE
RECONSTRUC
TION LEVY
TOTAL
MINING
1990 5.4% 3.6% - - 8.9%
1991 1.3% 4.9% - - 6.3%
1992 0.6% 6.1% - - 6.7%
1993 3.9% 6.6% 2.3% - 12.8%
1994 4.3% 7.7% 2.9% - 14.9%
1995 7.4% 7.6% 2.9% - 17.9%
1996 2.2% 8.4% 4.0% - 14.5%
1997 1.6% 5.7% 4.1% - 11.5%
1998 1.8% 6.3% 3.9% - 12.1%
1999 3.5% 5.4% 3.1% - 11.9%
2000 1.1% 8.4% 4.2% - 13.7%
2001 1.3% 6.5% 3.9% 0.2% 11.9%
2002 0.9% 5.6% 3.7% 1.0% 11.1%
2003 1.8% 5.1% 3.7% 0.4% 11.0%
2004 1.9% 4.0% 2.5% 0.7% 9.1%
2005 4.2% 3.7% 3.0% 0.4% 11.2%
2006 5.5% 4.3% 3.0% 0.2% 10.2%
2007 5.3% 4.5% 3.8% - 13.7%
2008 6.0% 4.8% 3.9% - 14.7%
2009 7.2% 5.2% 6.0% - 18.4%
2010 9.9% 5.9% 5.4% - 21.3%
Source: Ghana Revenue Authority, Minerals Commission of Ghana
The table below shows the percentage shares of mining tax revenues to total GRA
collections.
Table 1.1: Percentage shares of mining tax revenue to total GRA collections.
While royalty payments and Pay As You Earn have increased consistently from 0.75 and
0.26 million Ghana Cedis in 1993 to 144.70 and 132.47 million Ghana Cedis in 2010
respectively, corporate tax does not show a particular pattern in the 1990s but is quite
consistent in the 2000s when it increased from 1.58 million Ghana Cedis in 2000 to 241.58
million Ghana Cedis in 2010.
University of Ghana http://ugspace.ug.edu.gh
27
Average annual growth rate the total GRA collections for the period under review amounts
to 36.7 percent whereas that of mining contribution amounts to 47.4 percent, a difference
of 107 basis points.
Another source of revenue to the government is dividend payments. The government of
Ghana holds not less than 10 percent in all mining companies and receives dividends from
the profits made. However, since 2000, the government has not received any dividends
from the mining sector.
2.5.3 Output Trends
Small scale and artisanal mining and medium/large scale mining are the differing
classifications of mining in Ghana. The medium/large scale was mainly state-owned
before the ERP and developed thereafter. However, these large scale mines are now
mainly owned by foreigners with Government of Ghana having a free-carried interest of
10% stake in all of the mines. For example, Ashanti Goldfields has about 82% foreign
ownership.
University of Ghana http://ugspace.ug.edu.gh
28
Below is a table of output of minerals in from 1990 to 2010.
Table 2.2: Production of minerals
Year Gold Diamond Bauxite Manganese
(1000 Oz) (1000 Cts) (1000 MT) (1000 MT)
1990 522.5 484.9 381.4 364.4
1991 946.3 702.2 352.9 326.0
1992 1,006.9 596.2 338.2 353.5
1993 1,251.0 584.8 423.7 294.8
1994 1,396.9 746.9 426.1 272.0
1995 1,630.3 627.3 513.0 245.4
1996 1,550.8 714.7 473.2 161.7
1997 1,644.6 698.6 504.4 273.2
1998 2,353.0 823.1 442.5 348.4
1999 2,257.7 680.3 355.3 638.9
2000 2,315.0 627.0 503.8 638.9
2001 2,205.5 870.5 715.5 1,212.3
2002 2,115.2 924.6 647.2 1,132.0
2003 2,208.2 927.0 494.7 1,509.4
2004 1,794.5 911.8 498.1 1,593.8
2005 2,149.4 1,062.9 726.6 1,714.8
2006 2,244.7 970.8 885.8 1,658.7
2007 2,486.8 837.6 748.2 1,156.3
2008 2,586.0 598.0 694.0 1,089.0
2009 2,930.3 354.4 490.4 1,012.9
2010 2,970.1 308.7 512.2 1,194.1
Source: Minerals Commission, Ghana.
On the GDP front, one cannot say Ghana is dependent on the mining sector. This is
because the mining and quarrying subsector contributions to GDP (measured in 1993
constant prices) have increased from 4.8% in 1990 to 5.7% in 1998 to 6.27% in 2009.
University of Ghana http://ugspace.ug.edu.gh
29
2.5.4 Export Earnings
According to Aryee (2001), the development of mining up to the late 1990s heightened
investor interest in the activity relative to other major economic sectors. The significance
of minerals as a contributor to the foreign exchange resources of the country is material
with gold as the flagship of the sector, typically accounting for some 95% of all mineral
earnings (Bank of Ghana, 2010). In 1983, total mineral exports accounted for 27.7% of
total exports whiles gold returned 93.76% of total mineral exports. In 1990, the share of
mineral exports in total exports declined to 23.0% whereas the share of gold to total
mineral exports also declined to 83.20. But in 1998, these shares increased to 39.59% and
95.81% respectively. As of 2009, mineral share in total exports was 33.5% whereas gold
exports share of total mineral exports was 97.4%.
Table 2.3: Mineral and Total exports for selected years. (Export value in million
USD)
Year Gold Exports Total Mineral
Exports
Total exports
1983 114.1 121.7 439.0
1990 201.6 185.9 808.2
1998 687.76 717.8 1813.2
2009 2551 2619 7809.4
Source: Bank of Ghana
University of Ghana http://ugspace.ug.edu.gh
30
2.6 Measures Adopted to Enhance National Revenue Mobilization
Tax administration effectiveness is necessary to attain the objectives of every reform of the
tax structure. The focus of the tax reforms after 1985 was broadened to include the
enhancement of efficiency of the tax administration and equity of the tax system. The pace
of the tax reforms during this period was also increased and more attention was paid to the
strengthening of private sector incentives. Until 1986, the primary function of tax
administration was facilitating, monitoring taxpayer compliance and preventing non-
compliance had operated very inefficiently.
The CEPS and IRS were unable to plan or carry out reforms to improve revenue
collection. As a major part of the tax reform programme, a considerable number of steps
were taken to strengthen the role of the revenue institutions in achieving the objectives of
increasing revenue and changing the structure of the tax system to make it more efficient
and equitable.
The most dynamic step taken was the conversion of the IRS and CEPS into autonomous
Services with new organizational structures in line with the state-owned enterprises in
1986. At the same time, the remuneration, career path and other human resource issues of
the new structures were reviewed and embodied in a collective work agreement. The
agreement also established new incentive policies including promotion procedures and
training for the staff of the IRS and CEPS. These initiatives were aimed at generating from
within the new organizations the necessary changes to improve productivity, particularly
in the area of tax collection.
University of Ghana http://ugspace.ug.edu.gh
31
Between 1986 and 1992, the IRS itself operated with Ministerial powers alongside the
Ministry of Finance. In 1992, this full autonomy was partially reversed, and currently IRS
is headed by a commissioner who reports to the Minister of Finance. The IRS however
continues to have a supervisory control over the collection of revenue in the country. In
terms of the functional reorganization of the revenue institutions, both revenue and
expenditure accounting have been detached from the Treasury and coordinated only by
regular returns submitted by the IRS and CEPS to the Secretariat, the Treasury and the
Central Bank (Terkper, 1994).
To facilitate trade by modernizing operations and clearance procedures at the ports and
harbours, all CEPS offices at the airport and harbours were automated and operating and
processing all imports and some exports through the Ghana Customs Management System
(GCMS).
There was also the introduction of scanning facilities at various points of entry into the
country. This was to ensure proper assessment of volume and content of containers and
packages. Tema and Accra were used as starting points on pilot basis.
The government made efforts to widen the tax net and minimise the incidence of tax
evasion through the review of exemptions and the zero rated items list. The VAT bill was
presented to parliament, completion of preparatory works for the issue of new taxpayers‘
identification numbers and provision for tax courts were example of such efforts. A new
system of collection of tax on lotteries also began in September 1997.
Renovation and rehabilitation works started on offices and residential houses of the
custom‘s administration to enhance administrative effectiveness. Computerization of the
University of Ghana http://ugspace.ug.edu.gh
32
Long Room at Tema was duly completed in 1999. A review of clearance procedures by
CEPS was done to reduce multiple state agencies involved in clearing of imports.
Decentralisation of clearing was to reduce unnecessary documentation and CEPS instituted
stiffer penalties for breach of discipline and misconduct among its workers.
The implementation of the WTO agreement on customs valuation and destination
inspection systems of imported goods began on April 1, 2000. Under this system, imports
irrespective of their mode of financing were subjected to destination inspection by
accredited inspection agencies unless specifically exempted by the Minister of Trade and
Industry. Again, there was the introduction of scanning facilities at the various shipping
ports and airports in Tema, Sekondi-Takoradi and Accra on pilot basis for proper
assessment of volume and content of containers and packages.
Work began on GCMS and Ghana community trade network to link major trade and
customs stake holders comprising Ministry of Finance, Ministry of Trade and Industry,
Freight forwarders, Ghana Ports and Highway Authority, Civil Aviation etc to an
information sharing pool, with the view to facilitating automation of custom procedures. A
computerisation of various state warehouses to facilitate easy monitoring was completed in
2000.
The implementation of the uniform Tax Identification Number (TIN) scheme by all
revenue agencies started in 2002. The LTU was created to facilitate the assessment,
processing and audit of the tax liabilities of major tax payers. There was also a reduction in
the period permitted for carrying over accumulated capital allowances, carryover of losses,
disclosure of foreign exchange gains and losses, disclosure of acquisition of depreciable
University of Ghana http://ugspace.ug.edu.gh
33
assets and increased investigation by the IRS. The range of goods exempted from import
duty and also those cleared from the ports on concessional terms was narrowed.
To minimise the charges imposed by banks on their customers, it was decided to remove
the Ghana community network (GCNET) processing for on importation of currency notes
by the banks.
The final effort was the integration of the three tax revenue agencies into one revenue
agency. The CEPS, IRS and VATS and the Revenue Agencies Governing Board (RAGB)
Secretariat were merged in accordance with Ghana Revenue Authority Act 2009, Act 791.
The GRA thus replaced the revenue agencies in the administration of taxes and customs
duties in the country. Its establishment was a culmination of years of plans to streamline
the administration of tax collection in Ghana which began in 1986.
2.7 Conclusion
The above observations show a particular trend of growth in the mining sector of Ghana. It
is also obvious that the Ghana Revenue Agency has come a long way to ensuring revenue
administration is effective. The fiscal trend of the mining sector also gives in-depth
knowledge on how and why the tax reforms were initiated.
University of Ghana http://ugspace.ug.edu.gh
34
CHAPTER THREE
REVIEW OF LITERATURE
3.0 Introduction
This chapter addresses an overview of theoretical and empirical literature on tax reforms
and revenue productivity with its related issues. The section also looks at the theory of
optimal taxation and defines the importance of tax buoyancy and elasticity. The section on
empirical literature looks at the different methods that have been used to estimate tax
buoyancy and elasticity by different researchers. Lessons from empirical literature are also
discussed.
3.1 Role of Taxation
It is only appropriate that a discussion of issues in tax reforms should begin with an
assessment of the role of taxation as a macroeconomic tool. In the past, tax systems were
used in developing countries to serve multiple objectives. These included, in addition to
mobilization of resources to finance government expenditure, promoting savings and
investment, inducing savings in particular forms to facilitate the process of channelling
savings into investment, directing investment into desirable activities, bringing about
greater equity in the distribution of income and correcting externalities. Taxation was
viewed as a powerful policy instrument to serve the dual role of securing resource transfers
University of Ghana http://ugspace.ug.edu.gh
35
to the public sector for application to planned uses and inducing the private sector to
operate in conformity with planned objectives (Islam, 2001).
Over the last three decades however, the view that has gained wide acceptance is that the
tax system should be assigned a much narrower role, i.e., it should focus on raising
revenue. According to the IMF, taxation creates distortions and the main object of tax
policy is to design a system that raises enough revenue to meet a government‘s revenue
target while minimizing the level of associated distortions (Mackenzie et al, 1997).
3.2 Theoretical aspects of tax reforms
Governments that embark on tax reforms aim to achieve many goals. Raising the
productivity of the tax system is one of the most important. To facilitate compliance and
collection, a tax system must be administratively feasible. In other words, it should be
simple for the same reason but also as an aim in itself, it must spread its burden equitably
to avoid misallocation of resources. It must not also upset the pattern of production, trade,
consumption, savings and investment. This justifies the assertion that tax reforms are
simply a matter of trade offs as the aims outlined above cannot be satisfied simultaneously.
Most countries that have embarked on tax reforms have aimed at raising productivity of
the tax system. UtunWai (1962) argues that tax revenue is very important because in the
long run revenue cannot lag behind expenditure. So unless public expenditure is expected
to grow at the same rate as national income, the government should ideally choose tax
bases that will expand proportionately with spending not GDP.
University of Ghana http://ugspace.ug.edu.gh
36
According to Mansfield (1972), the success of tax reforms in raising revenue productivity
can be evaluated by looking at the buoyancy and elasticity of the tax system. Buoyancy
measures the automatic effects of discretionary changes in the tax system to changes in
income whereas elasticity is built in responsiveness of tax revenue to movements in
national income abstracting from the effects of discretionary changes in tax system.
Choudhry (1979) complements the work of Mansfield (1972) by suggesting that the size of
buoyancy and elasticity is influenced by many factors. Firstly, the composition of bases
influences the size of elasticity and buoyancy of a tax system. Some components of the tax
base may be primarily responsible for lowering or raising the overall buoyancy and the
elasticity of tax revenue. Secondly, the progressive elements in the tax system also
influence the size of the elasticity. When per capita income rises the whole population
moves up in the income scale, those who were paying taxes either directly or indirectly
will pay larger taxes and some of those who were below the exemption level of income
will start paying. The progressive element in the tax system also gives it some degree of
built-in protection against inflation. When prices rise, money incomes increases and the
tax system cause the government revenue to increase at least with national income and
very rapidly if the system is sufficiently progressive. Thirdly, better tax administration and
prevention of evasion and avoidance as well as elements of tax policy such as using ad-
valorem rates to increase revenue as the value of the base rises, influences the size of
elasticity and buoyancy.
The size of elasticities and buoyancies are believed to be affected by other factors which
are not considered explicitly (Wilford, 1978). For instance if the government spends most
University of Ghana http://ugspace.ug.edu.gh
37
of its revenue on defence or something from which the recipients does not consider he
derives an income, there will no longer be income effects on tax revenue.
The distribution of income also influences the size of the elasticities, for example if there
is a shift of income towards higher income groups and the tax system is progressive, the
size of buoyancy and elasticity will increase.
In addition, the industrial origin of output influences the size of elasticity and buoyancy.
As the share of manufactured output as a ratio of GDP increase tax revenue increases
because it is easy manufactured output since there will be few centres (factories) to collect
the tax.
Inflation is also another factor that influences the size of elasticity and buoyancy because it
pushes people into higher tax brackets and tends to overstate company profits and incomes
of individuals.
Studies by Bryne (1983), Baas (1974) and Tanzi (1988) show that discretional measures
(legislative enactment, improvement in tax administration, changes in tax rates and bases,
etc) and economic growth affect one or more of the factors that influence the buoyancy.
The effect of such factors on buoyancy is in turn felt in the tax ratio (tax revenue relative
to GDP).
University of Ghana http://ugspace.ug.edu.gh
38
3.2.1 Tax Reform and Optimal Taxation
The literature on tax reform has been growing rapidly suggesting the theoretical and
practical importance of the subject. Interestingly most of the literature has been more
descriptive than analytical. The techniques applied to evaluating success or failures of tax
reforms are not well documented. Normally the analysis of tax reform has tended to focus
on evaluating the objectives of those reforms: revenue adequacy, economic efficiency,
equity and simplicity. According to McMahon and Berrios (1991), the need for tax reform
arises from the deficiency of the existing tax system in achieving these objectives.
Revenue adequacy is the basic elementary standard that a tax system ought to achieve. The
existing budget deficits in many developing countries like Ghana seem to suggest that the
tax systems are not revenue productive. Some may overlook this and attribute the cause of
deficits to excessive spending, or temporarily adverse economic conditions. In situation
where budget deficits persist for long periods increased revenue should be the main
objective of tax reform. Obviously, this is bound to depend on the circumstances of each
country. Certainly, few, if any, developing countries can afford to adopt tax reforms, no
matter how desirable they might be on other grounds, if they lead to substantial revenue
losses. Goode (1987) argues that it is hard to gain serious consideration for any revenue-
neutral reform proposal. By and large, in Africa most governments consider revenue gain
as the primary motive for tax reform.
Optimal tax theory still has a significant influence on academic research in tax reform.
Newberry and Stern (1987) applied a normative framework to analyze a tax reform
process. The optimal taxation approach emphasizes the need to analyse the impact of tax
University of Ghana http://ugspace.ug.edu.gh
39
reform and evaluate both its administrative costs and its effect on social welfare. This
framework has been criticised for its inability to identify the real practical needs of tax
reform in developing countries. Its first major shortcoming is that it requires substantial
data, which are scarce or non-existent in many developing countries.
Secondly optimal taxation has assumed the existence of perfect administration. However,
recent reform experiences have revealed a serious lack of administration capacity in
virtually all developing countries, showing the need for simpler administrative structures.
The impact of optimal taxation on tax reform in developing countries has been small and
indirect because of this deficiency (Gillis, 1989). In addition, under the optimal taxation
approach the analysis of revenue productivity of a tax would be of less significance.
The theories of optimal taxation are analogous to the examination of the principles of
taxation, where lump-sum taxes are impossible. The analysis of optimal commodity
taxation began with Ramsey (1927) but the subject expanded in the 1970s following
Diamond-Mirrelees papers of 1971.
The general principles of optimal taxation can be summarized as follows:
Tax revenue is raised most efficiently by taxing goods or factors with inelastic
demand or supply and
Taxation concerned with distribution and with externalities or market failures
should as much as possible address the problem.
Thus for distribution, one should look for the sources of inequality and should concentrate
taxation there. In the case of externalities, one should attempt to tax (in the case of
University of Ghana http://ugspace.ug.edu.gh
40
negative externality) or to subsidize (in the case of positive externality) directly the good
or activity that produces the externality (Stern, 1988).
In cases where equity and efficiency must be balanced, as in the design of direct and
indirect taxes on consumers, the tax rates will depend on the exact form of the social
welfare function (Newbery, 1988).
By ‗tax reform‘, we mean a movement away from some given status quo. According to the
general theory of tax reform, it will be beneficial to switch taxation at the margin from i to
j, if the marginal cost of tax i exceeds that of tax j. More generally, a tax reform is
beneficial if it increases both revenue and social welfare.
In relation to shadow prices, a tax should be increased if the direct impact on households
of making the change exceeds the cost at shadow prices of the extra demand generated.
The shadow price of a good embodies the welfare consequences of the general equilibrium
adjustments that flow from an extra demand for that good. Thus the shadow price depends
on the way in which the economy adjusts (Stern, 1988).
Although the reform suggested by optimal taxation are based upon rigorous economic
theory as Thirsk (1995), Deaton (1988) and McClure (1989) emphasized, putting them into
operation leads to as intractably large number of rates, which would be difficult to
calculate and infeasible to administer effectively. Slemrod (1990) has argued that the
optimal tax theory can serve as a guide to designing ‗optimal tax systems‘ only if one
considers the technology of tax collection, i.e., the feasibility of tax instruments, the cost
of administration and compliance.
University of Ghana http://ugspace.ug.edu.gh
41
According Thirsk (1995), despite flirtation with the undeniable logic of optimal tax theory,
many tax reforms have focused on achieving broader based taxation at more uniform rates.
Unlike the earlier pleas for broader based taxation, which called for greater progressivity,
the current appeals are for lower marginal tax rates and reductions in the level of tax-
induced distortions that are prompted by high rates.
According to Chhibberet al (1988), experience with growth-oriented adjustments
programmes in developing countries indicates that tax reform is an essential component of
any comprehensive strategy for structural adjustment and resumption of growth.
As Musgrave (1988) had pointed out, however, tax reform in developing countries
involves broad issues of economic policy as well as specific problems of tax structure
design and administration.
First, there are the central problems of revenue requirements and how to fit the revenue
structure into development policy. This area of concern includes the impact of alternative
taxes on saving and investment and their implications for the macro balance (domestic and
foreign) of the economy.
Second, there is the important goal of securing a fair distribution of the tax burden. Among
the more specific tax issues, attention needs to be given to the composition of the tax
structure as well as the design of its major components. The problem throughout is not
simply to determine what would be desirable but also to assess what is administratively
practicable and within the ballpark of political feasibility.
University of Ghana http://ugspace.ug.edu.gh
42
According to Kusi (1998), a common feature of the tax structures in most developing
countries is that they are complex (difficult to administer and comply with), inelastic (non-
responsive to growth and discretionary policy measures), inefficient (raise little revenue
but introduce serious economic distortions), inequitable (treat individuals and businesses
in similar circumstances differently) and unfair (tax administration and enforcement are
selective and skewed in favour of those with the resources to defeat the system). There is a
heavy reliance on taxes on international trade, which undermines long-term international
competiveness. User charges and taxes on income and property contribute only a small
proportion of total revenues. Agricultural income, fringe benefits, wealth bequests, land
and property exist in theory but have rendered ineffective by design problems or the lack
of administration, or both, while personal and corporate income taxes are levied on narrow
bases at high rates. Sales taxes are levied in a cascading manner, thereby pyramiding and
in some cases, more than hundred percent full forward shifting (Khalilzadeh-Shirazi and
Shah, 1995; Bird, 1995; Shalizi and Squire, 1988).
Faced with mounting budget deficits and having to cut expenditures as far as is prudently
possible, particularly on public investment and social spending, a number of developing
countries have undertaken to restructure their system of taxation to seek higher revenue or
improve the revenue elasticity and buoyancy of the tax system. Other goals of these reform
processes have been to eliminate the disincentive effects of the levels of taxation; to reduce
the economic inefficiencies induced by the distortionary taxation of assets and sectors; to
protect the poorest of the poor from the tax net; and to provide partial relief from the
unwelcome effects of inflation (Khalilzadeh-Shirazi and Shah, 1995). Thus revenue
University of Ghana http://ugspace.ug.edu.gh
43
enhancement, economic efficiency, horizontal equity and simplicity issues have dominated
the world agenda on tax reform.
Despite the wave of tax reforms over the globe, and the remarkable similarity in the broad
directions of the reforms, a number of unresolved and controversial issues remain. For
example, vertical equity and international income taxation have received only scanty
attention in many tax reforms. Emphasis on redistributive role of the tax system is
gradually waning - a direct consequence of the fact that tax evasion is so pervasive. The
proper role of progressive income taxation is an extensively debated issue, as is the
question of whether personal income tax should have fewer brackets and rates on account
of simplicity. Although progressivity remains high on the political agenda in theory, often
the political will to enforce income tax compliance (expenditure) or cash-flow taxes,
perplexing philosophical and transitional issues continue to dominate current discussions.
Vertical equity is increasingly being perceived as an elusive goal and therefore is being
assigned a lower order of priority in tax reform. All recent attempts at tax reform have
curtailed tax preferences, especially for investment, but some economists argue that certain
incentives, such as investment tax credit, are desirable because they lower user cost of
(new) capital, thereby encouraging greater capital formation.
3.2.2 Measuring Tax Revenue Productivity
Hindrichs (1966) and Musgrave (1969) expressed the need or the importance of ―tax
handle‖ in determining ―tax efforts‖. The four (4) main approaches to assess the tax
University of Ghana http://ugspace.ug.edu.gh
44
performance of a country are (i) ―ability to give up approach‖, (ii) ―efficient resource use
approach‖, (iii) ―ability to collect approach‖ and (iv) ―comparison with average
performance approach‖ which is known as the ―stochastic approach‖ Musgrave (1969).
The commonly used approach for measuring tax effort; is to regress the tax ratio on a set
of variables that serves as proxies for ―tax handles‖ of a country. The set of variables
include the major determinants of output in a country (Bahl, 1971) and (Chelliah, 1971).
This can be expressed in a functional form as: T/Y=f(V), where, T is the tax revenue, Y is
the GDP or GNP, T/Y is the tax ratio and V is the vector of ‗tax handles‘. The equation
provides an average relationship between the tax ratio and the set of explanatory variables
chosen. For example in an IMF study, the tax ratio (T/Y) was regressed on per capita non-
export GNP (in U.S Dollars), export ratio including mineral exports, and the share of
mining in GDP. The predicted tax ratio therefore gives the ratio that the country would
have if it had made the average tax effort. Thus it becomes a measure of the taxable
capacity of the country while the regression coefficients act as the ‗average‘ effective rates
on the base. According to Bird (1976), the successful measurement of taxable capacity
used in studies depends critically on the a priori justification of the explanatory variables
as affecting only taxable capacity and not at all either demands for higher public
expenditures or willingness to tax.
Another method of assessing tax performance is via tax buoyancy and elasticity. The
buoyancy of tax can be estimated in two ways: (i) by calculating the ratio of percentage
change in tax revenue to percentage change in national income/GDP, or (ii) by a
regression of the tax revenue on the tax base (e.g. national income/GDP) after taking the
natural logarithm for each of them. It measures the relationship between the proportional
University of Ghana http://ugspace.ug.edu.gh
45
change in tax revenue and those of the tax base and therefore, gives a quantitative measure
of the effectiveness of tax policy in terms of stimulating public resources (Teera, 2002).
The tax effort approach to measuring tax performance is termed ―static‖ since it gives the
potential for tax increase in a country at a given point of time through comparisons with
other countries. However, in order to determine if a country has made efforts at increasing
tax revenue over a period, tax performance in the dynamic sense is important to know. The
tax buoyancy provides such a dynamic index of tax performance, which measures the
sensitivity and response of the tax system with respect to income/GDP.The buoyancy of a
tax system reflects the total response of tax revenue to changes in national income as well
as discretionary changes in tax policies over time.
The elasticity of a tax system though closely related to buoyancy measures the
responsiveness of tax revenue to changes in national income abstracting from discretionary
changes in the tax structure. Hence, in estimating the elasticity of the tax system, one must
correct for effects of discretionary changes in tax policy on historical tax revenue series. It
is important to obtain both the buoyancy and elasticity of the tax system because the
responsiveness of tax revenue to changes in Gross Domestic Product (GDP) is of two
types: the automatic response to GDP changes, and the response resulting from
discretionary changes in tax rates, tax base and tax legislation. The combined effect of the
two is known as the buoyancy of a tax. Buoyancy is measured by generally estimating the
responsiveness of tax revenue to changes in income or output without taking into account
discretionary changes in tax policy. According to Mansfield (1972) and Osoro (1993), the
traditional method of measuring tax buoyancy starts with the following model:
from which the double log function is derived, where T is
University of Ghana http://ugspace.ug.edu.gh
46
historical tax revenue, Y is GDP, and β is the estimated buoyancy, α is the constant term
and Ɛ is a stochastic disturbance term. Tax elasticity is measured in the similar way as
buoyancy with the exception that in elasticity estimation discretionary changes in the tax
policy are taken into account.
Discretionary changes in the tax policy include changes in the tax rate and/or base;
changes in the efficiency of tax administration; the introduction of new taxes and the
abolition of some others. This therefore means that historical tax revenue series have to be
refined by adjusting them to exclude revenue changes attributable to discretionary
measures.
Estimating the income tax elasticity is useful for displaying the extent to which tax system
is responsive to changes in the tax composition and the value of GDP (Teera, 2002). When
the elasticity of major revenue sources remains low despite tax reforms either due to low
base, or evasion or avoidance, the government raises additional resources through
discretionary measures. Then, the growth of tax revenue comes through high buoyancy
rather than through elasticity. The coefficient of elasticity depends on the level of tax base
to changes in income. This makes it possible to break up the value of elasticity into two
components: the response of the tax base to a change in income, and the response of tax
yield to a change in the tax base of individual taxes through decomposition of elasticity
(Musgrave, 1969). The value of base-to-income elasticity does not depend on the
progressivity of tax rates; it simply relates the responsiveness of the tax-base to a change
in income. The growth of the base depends on the way the structure of the economy
changes with economic growth. The tax-to-base elasticity depends on the tax rate; if the
rate structure is progressive or if there is an improvement in tax administration; the tax-to-
University of Ghana http://ugspace.ug.edu.gh
47
base elasticity will be raised by preventing evasion. The decomposition of elasticity in this
manner permits us to identify the source of growth of tax revenues.
The major methods used include proportional adjustment methods (Baas and Duxin,
1972),(Mansfield, 1972), (Osoro, 1993), (Kusi, 1998), constant rate structure (Choudhry,
1975), (Tanzi, 1976), dummy variable technique (Wilford, 1998), (Hyuha and Dumba,
1993) and the divisia index (Choudhry, 1979). However, a complete adjustment of
historical revenue series is not possible in any of the methods.
The first method (Proportional Adjustment) requires the use of budget estimates of tax
yields resulting from discretionary changes. Not only are such data difficult to obtain, their
reliability is questionable as the actual discretionary outcomes may differ significantly
from the changes proposed in the budget. The data on discretionary revenues provided by
the Revenue Agencies are not only incomplete but suffer from these two limitations
among others. Firstly, some of the measures proposed at the time of budget-making were
never implemented, and therefore the financial effect could be different than reported in
the data provided. Secondly, the calculation of financial effect of tax measures is based on
the ‗static‘ model rather than dynamic, therefore ex-ante and ex-post differential can exist.
Proportional adjustment method involves subtracting budget estimates of tax yield owing
to discretionary changes form the actual yield. Multiplying it by a sequence of
multiplicative factors then further refines the adjusted revenue data.
The second method (constant rate structure), is not used very commonly because it places
heavy demands on the availability of data, for instance it requires data on effective tax
rates and the changing composition of availability of data (Anderson, 1973). Provided
University of Ghana http://ugspace.ug.edu.gh
48
these data are available and both the tax and its base are defined narrowly enough to
permit application of the reference year rates to later year tax bases, this method generates
the most accurate revenue series. However, such data are hard to come by particularly in
the less developed countries. Another limitation of this method is that, the elasticities of
individual taxes and the overall tax system estimated by using this method as an
adjustment technique depends on the variation of bases with respect to GDP
The dummy variable method does not require the use of disaggregated data on taxes, but it
cannot be used properly when discretionary tax changes are quite frequent in the past.
Furthermore, even if the discretionary changes are not very many, the specification of the
estimation equations can be problematic unless there is information on the nature of the tax
changes and the extent to which their effects are independent of one another. The dummy
variable technique involves introducing a dummy variable for each exogenous variable
that significantly influences revenue generation. However, this method is not appropriate
when discretionary changes have been made more frequently in the past.
The Divisia Index method uses only historic data and does not require the collection of
specific information on revenue effects of discretionary tax changes, or on the frequency
of past discretionary tax changes. According to Gillani (1986), the Divisia index approach
of measuring the revenue effects of discretionary is widely used for measuring technical
change and isolating the effect of exogenous factor on a variable. This method is expected
to provide a reasonable measure of the effects of discretionary changes especially when
information on revenue effects of discretionary changes is not available. However, apart
from the questionability of using time trends as proxies for discretionary measures, this
approach leads to over or under estimation depending on whether discretionary changes
University of Ghana http://ugspace.ug.edu.gh
49
produce positive or negative income effect. The use of divisia index also does not allow
the estimation of elasticities of components of tax revenue because it is assumed that each
base represents its own tax category (Choundry, 1979). However, two caveats may be
noted: first, it may underestimate or over-estimate the revenue effects of discretionary tax
change; and secondly, in case of large revenue effects, it may give unsatisfactory results.
3.3 Tax Types
This section reviews theoretical underpinnings of mining sector taxes considered in this
study.
3.3.1 Royalties
According to Cairns (1985), royalty is a payment exacted by the owner of a resource. Most
governments of natural resource endowed economies have used the royalty system to
mobilise revenue from natural resource operating companies. Royalties are an attractive
source of revenue to the government because they are easier to administer than many other
fiscal instruments (Sunley and Baunsgaard, 2001). Moreover, royalties ensure that
companies make a minimum payment for the minerals they extract. Otto (2000) asserts
that over the past century, there has been a trend to de-emphasize tax systems based on
royalties and to switch to tax mechanisms that are profit based. However, many nations
still retain royalty taxes. Royalty taxes can be in various forms; revenue based, profit
based, unit of production, or an ad-valorem basis (Sarma and Naresh, 2001).
University of Ghana http://ugspace.ug.edu.gh
50
For the case of unit of production, the base of the tax is the quantity produced which is
easy to check. In other words, a unit royalty is a levy applied to the per unit volume of
production. In the case of ad-valorem type, on the other hand, the tax is imposed based on
the price of the mineral. Strictly speaking, the ad-valorem tax is imposed on the value of
the oil or mineral extracted. However, when the true market price is not known, some
formula(e) can be used for determining the value for tax purposes from the quantity
produced (Sunley et al, 2002).
Revenue based royalty approach tax revenue with no account for costs of production.
According to Emerson and Lloyd (1983), revenue based royalty system discourages
production in the natural resources sector. In other words, since royalties tend to raise the
marginal cost of extracting mineral ores, they can deter investors if imposed at high levels.
As a result, the sliding scale royalty has been introduced to help reduce distortions in
mineral exploration. In line with the sliding scale, the R-factor is used to make allowances
for costs when levying royalty. It is calculated by multiplying the royalty rate by the value
of mineral at the valuation point which takes into account all cost and revenue of the
project to the day of calculation. Mathematically, R = X/Y, where X and Y are the
cumulative revenue and cumulative expenditure respectively (Quiroz, 2003).
According to Sinner et al (2005), profit-based royalty system is a mechanism whereby the
resource owner receives a share of the profits once all significant costs have been
recovered.
University of Ghana http://ugspace.ug.edu.gh
51
3.3.2 Company Income Tax
Company tax is a tax levied on the profits of all companies and it is calculated after
interest and all revenue allowances such as capital allowances have been catered for, but
before dividend distribution (Bannock et al, 1998). In most economies, company income
tax treats mining sectors companies more favourably than other sectors. The intention is to
attract investment into the natural resources sector. For example, minerals require huge
capital outlay for their extraction. Therefore, to attract investment, the tax system applied
must be more favourable (Otto, 2000). This has led to many countries including Ghana, to
allow exploration and development costs to be deducted for taxes purposes in the mining
industry.
Corporate income tax many comprise the basic rate structure, depreciation tax,
supplementary levies, tax incentives and withholding provisions (Sarma and Naresh,
2000). Various governments also allow for accelerated depreciation allowances as a way
of promoting the development of the mining industry. There are many approaches to the
calculation of the annual value of depreciation including the straight line method, reducing
balance method, sum of the years digit, annual capital charge. Overall corporate income
tax rates in several countries ranges from 25% to 35%.
University of Ghana http://ugspace.ug.edu.gh
52
3.3.3 Export and Import Taxes
Because of the capital intensive nature of the mining sector, most countries exclude the
industry from paying import duties. This is done to promote investor confidence into the
sector. Again, most mining companies depend on imported equipment for exploration,
development and operational activities. One major setback of the import tax is that most
governments have observed that it directly and negatively impacts on the competitiveness
of their mining industries on the global market. In Indonesia, import duty on foreign
equipment is excluded for the first ten years, and then capital equipment and spare are
taxed at around 20%. Chile, Kazakhstan and Mexico impose 10%, 11% and 20% of import
tax respectively. Export duties are imposed on raw minerals. In addition to acting as a
general revenue raising device, export duty is used as a means to encourage value addition
to the raw material produced. With increased global competition, most nations have zero
rated export duties on minerals regardless of the degree of processing, (Otto, 2000).
Ghana‘s mineral industry for example is exempt from import duties (but for a large mine
there may be exceptions) and export duties are zero-rated.
3.4 Empirical Literature Review
The theoretical and practical importance of tax reforms attracted the attention of several
researchers. This resulted in studies being carried out to examine the effect of discretionary
changes on revenue productivity. A review of the empirical literature below is organized
University of Ghana http://ugspace.ug.edu.gh
53
on the basis of method used in adjusting tax revenues in trying to draw lessons for the
present study.
Few studies have been undertaken using the Divisia Index approach. The work of
Choudhry (1979) is particularly appealing as the impact of discretionary changes on
revenue productivity in USA, UK, Malaysia and Kenya was his focal point of empirical
inquiry. The empirical results indicate that the estimated elasticities and buoyancies differ
considerably between and within pairs of developed and developing countries. The
estimates for the USA and UK were smaller than those for Malaysia and Kenya. The
estimated buoyancy for USA was 1.04, UK 1.18, Malaysia 1.7 and Kenya was 1.42, also
the estimates of elasticity ranges from a lowest of 1.04 for the USA to high of 1.57 for
Malaysia. This suggests that developing countries, which are experiencing growth, exhibit
a more growth elastic revenue base than the developed countries. The findings of this
study show that the discretionary tax measures generated additional tax revenue. Gillani
(1986) also used the Divisia Index method to estimate the elasticity and buoyancies of
federal taxes in Pakistan for 1971-72 to 1982-83. The long run estimates were an elasticity
of 1.26 and buoyancy of 1.17. Short run estimates were however 0.83 for elasticity and
0.74 for buoyancy. Results of the study showed high elasticity coefficients for all taxes
except for export taxes.
Several authors have used proportional adjustment technique to adjust tax revenue. In this
regard, the work by Baas and Dixon (1972) is revealing. One point that emanated from this
study on U.K tax system was that although the discretionary measures reduce tax revenue,
they improved the progressivity of the individual income tax system. Their statistical
results indicate that individual income tax had a buoyancy of 1.02 and 1.42 for the period
University of Ghana http://ugspace.ug.edu.gh
54
1950/51-1970/71 and 1960/61-1970/71 respectively, while the elasticity estimates were
1.37 and 1.53 in the same periods. In another study, Mansfield (1972) indicates that the
size of both buoyancy and elasticity were favourably influenced by relatively high base to
income elasticity of major taxes. Other authors like Rao and Katsande (1985), Osoro
(1993) and Bryne (1983), reveal in their studies that discretionary changes generated
additional revenue.
Other authors employed the dummy variable technique. In this respect, the work by
Wilford (1978) points to the fact that tax reforms greatly influenced the tax systems of
Central America. The work of Hyuha and Dumba (1993) also points to the same findings.
Dummy variable technique which they employed to estimate elasticity was not appropriate
because of the many discretionary changes that took place in different years during the
study period.
Osoro (1993) examined the revenue productivity implications of tax reforms in Tanzania
for the period 1979 to 1989. In the study, the tax buoyancy was estimated using double
log form equation and tax revenue elasticity using the proportional adjustment method.
The result gave an overall elasticity of 0.76 with buoyancy of 1.06. The study concluded
that the tax reforms in Tanzania had failed to raise tax revenues. These results were
attributed to the government granting numerous tax exemptions and poor tax
administration within the sample period.
In another study, (Osoro, 1995) used the Prest (1962) method to estimate the individual
tax elasticity and that of the overall tax system in 1970-1980.He found that the elasticity
of the overall tax system declined from 0.85 in 1970 to 0.782 in 1980. Income tax and
University of Ghana http://ugspace.ug.edu.gh
55
Sales tax, which were elastic in 1970‘s, became inelastic in the 1980‘s. Import duty,
which was inelastic in the 1970s, became elastic in 1980s. He attributed these changes to
reduction in import duty rates and a rise in imports, rapid changes in tax base, stemming
from steep exchange rate depreciation.
Ariyo (1997) evaluated the productivity of the Nigerian tax system for the period 1970-
1990. The aim was to devise a reasonable accurate estimation of Nigeria‘s sustainable
revenue profile. In the study, tax buoyancy and tax revenue elasticity were estimated
using the double log form and the proportional adjustment method respectively. The slope
dummy equations were used for the oil boom and Structural Adjustment Programmes.
The study revealed an overall satisfactory tax productivity level but wide variations in the
level of tax revenue by various tax sources. The variations were attributed to the laxity in
administration of non-oil tax sources during the oil boom periods. The study further
asserted that there was need to improve the tax information system to enhance the
evaluation of its performance and facilitate adequate macro-economic planning and
implementation.
Chipeta (1998) evaluated effects of tax reforms on tax yields in Malawi for the period
1970 to 1994. The results indicated buoyancy of 0.95 and an elasticity of 0.6. The study
concluded that the tax bases had grown less rapidly than GDP.
Muriithi and Moyi (2003) applied the concepts of tax buoyancy and elasticity to
determine whether the tax reforms in Kenya achieved the objective of creating tax policies
that made yield of individual taxes responsive to changes in national income. They used
the double log equation to estimate the responsiveness of tax yield on income. The results
University of Ghana http://ugspace.ug.edu.gh
56
showed that tax reforms had a positive impact on the overall tax structure and on
individual tax handles. The study concluded that despite the positive impact, the reforms
failed to make VAT responsive to changes in income. The difficulty here that VAT had
been around for eleven years only and subjecting it alone in a regression model did not
make statistical sense.
Milton Ayoki, MariosOwona and Moses Ogwapus (2005) researched on tax reforms and
domestic revenue mobilization in Uganda by using the proportional adjustment method.
Their findings revealed that the reforms had a positive impact on direct taxes as tax-to-
income elasticity index grew from 0.706 to 2.082 after the reforms whiles direct taxes also
moved from 1.037 to 1.306. They concluded that the reform was necessary to the
economy but there was room for improvement.
Suliman (2005) used the dummy variable method to look at the impact of trade
liberalization on revenue mobilization and stability in Sudan. The result shows that, the
tax system as a whole is not buoyant or elastic and the same results were obtained for the
major tax handles. Overall elasticity was inelastic with an index of 0.82 whiles the
elasticities of the individual taxes were divergent with the following indexes: import duty
(0.83), excise tax (0.82), income tax (1.26) and profit tax (1.57). The conclusion was less
buoyant and elastic system provides an explanation for the low tax efforts and the
relatively low and declining government spending.
FaizBilquees (2004) studied the Elasticity and Buoyancy of the Tax System in Pakistan by
using the Divisia Index method over the 1974/75 to 2003/04 period. The results gave a
total tax buoyancy and elasticity after the reform as 0.92 and 0.88 respectively and
University of Ghana http://ugspace.ug.edu.gh
57
concluded that, overall the use of discretionary tax measures has been relied upon
significantly as a source of revenue augmentation in Pakistan.
In Ghana, Kusi (1998) studied tax reform and revenue productivity of Ghana for the
period 1970 to 1993. Results showed a pre-reform buoyancy of 0.72 and elasticity of 0.71
for the period 1970 to 1982. The period after reform, 1983 to 1993, showed increased
buoyancy of 1.29 and elasticity of 1.22. The low buoyancy and elasticity during the pre-
reform period was attributed to smuggling, unrecorded trade, tax evasion and laxity in tax
collection. The results also showed that income tax had the lowest elasticity and therefore
he recommended that authorities should move away from income based taxes in favour of
consumption taxation. The study concluded that the reforms had contributed significantly
to tax revenue productivity from 1983 to 1993.
Brafu-Insaidoo and Obeng (2008) studied the effect of import liberalization on Tariff
revenue in Ghana for the period 1966 to 2003. The study adopted the Singer (1968)
approach to estimate the duty buoyancy and elasticity for the study period and the
following results were obtained: for the entire period 1966-2003, buoyancy was 0.556 and
elasticity was 0.282, the period before import liberalization (19965-1982), gave buoyancy
of 0.33 and elasticity of 0.814, and for the period after the import liberalization (1983-
2003), buoyancy was 0.313 and elasticity was 0.049. A comparison of the results indicates
that duty buoyancy outweighed duty elasticity for the entire study period, meaning that
discretionary tax measures (DTMs) have improved tariff revenue mobilization over the
period. Overall evidence obtained however indicates that import tax is neither buoyant nor
elastic in Ghana, so the study suggestedthat a lot needs to be done to improve efficiency in
the customs collection administration system.
University of Ghana http://ugspace.ug.edu.gh
58
Twerefou, Fumey, Osei-Assibey and Asmah (2010) adopted the Dummy Variables
Approach to correct for the effects of Discretionary Tax Measures. Their findings
revealed that the overall tax system in Ghana was buoyant and elastic in the long run and
buoyancy exceeded the elasticity, but in the short run the reverse was the case. They
observed an improvement in both buoyancy and elasticity over the reform period (1985-
2007) as evidenced in pre-reform buoyancy and elasticity coefficient which were
generally less than unity but became greater than one after the reform. Overall tax
elasticity was estimated to be about 1.03, suggesting that the responsiveness of the tax
system to a unit change in GDP was more than unity.
This study differs from the previous studies conducted in Ghana in the sense that all the
earlier studies considered nationwide revenue productivity of the tax system. This study
seeks to measure the revenue productivity of a specific sector in the country. Elsewhere,
this study uses the dummy variable approach to measuring the tax buoyancy and elasticity
of the mining tax system.
University of Ghana http://ugspace.ug.edu.gh
59
Author Coverage Methodology Major Findings
Osoro (1993) Tanzania, 1979-1989 Proportional Adjustment Tax reforms failed to raise the tax revenues attributable to
numerous tax exemptions and poor tax administration
Moyi and Muriithi (2003) Kenya, 1973-1999 Proportional Adjustment Tax reform had positive impact on the overall tax structure and
on the individual tax handles
Bilquess (2004) Pakistan, 1974-2003 Divisia Index Use of discretionary tax measures has been relied upon
significantly as a source of revenue augmentation
Kargbo (2010) Sierra Leone 3SLS and Divisia Index Discretionary measures are statistically significant and that tax
reforms failed to increase the productivity of the tax system.
Agriculture and mining have negative impacts on tax revenue
sharesBaas and Dixon (1972) U.K, 1950-1971 Proportional Adjustment Discretionary measures reduced tax revenue but improved
progressivity of the individual tax system.
Ariyo (1997) Nigeria, 1970-1990 Proportional Adjustment Satisfactory tax productivity level but wide variations in the
level of tax revenue by various tax sources
Ayoki et al (2005) Uganda Proportional Adjustment Tax reform had a positive impact on direct taxes
Twerefou et al (2010) Ghana, 1985-2007 Dummy Variable Overall tax system is income elastic in the long run
Source: Author's own compilation
Table 2.4: Summary of Some Key Empirical Studies
3.4 Problems with tax reform
One problem of tax reform derives from the existence of an initial allocation. Though a
new tax system may be more efficient and more equitable than the existing one, the
transition from old to new may cause a redistribution of resources to occur than in itself is
undesirable. For example, it has often been suggested in the U.S. that the tax subsidy for
state and municipal bonds be removed. If this were done unexpectedly, it would cause a
capital loss for the holders of such bonds, but not for other, otherwise identical individuals.
Such treatment may be thought of as a violation of horizontal equity (Feldstein, 1976)
University of Ghana http://ugspace.ug.edu.gh
60
which may be explicitly accounted for in an expanded social welfare function (King,
1982). This problem undoubtedly is one of the reasons why tax reform is so difficult to
achieve.
A second general problem of tax reform is that the direction in which to move from the
current system is not always evident. Even if all distortions can be reduced somewhat, this
may not increase economic efficiency. The basic difficulty is that one can only be sure that
movement in the direction of a global optimum will improve matters if we are sufficiently
close to that optimum initially. A related problem is whether one can increase economic
efficiency in a piecemeal fashion, by remaining distortion, one at a time. In general, such a
scheme for tax reform may decrease welfare along the transition path to a global optimum.
According to Broadway and Harris (1977), the prevention of this requires extreme
restrictions on preferences and production sufficiency.
3.5 Lessons for Tax Reform
Tax reform experience to date offers some important insights into useful tax policy design
and institutional development. A detailed discussion of this issue is provided in
Khalilzadeh-Shirazi and Shah, (1995). A brief summary is presented below:
The value added tax should be an instrument of choice for developing countries
contemplating reform of their sales tax. According to Harberger (1990), there was
no such thing as value added tax (VAT) some fifty years ago. Since its introduction
in the early 1950s, however, VAT has become a fiscal innovation that has swept
University of Ghana http://ugspace.ug.edu.gh
61
half of the world, including many developing countries. The VAT has become an
instrument of choice for most developing countries contemplating reform of their
sales tax. VAT can provide greater revenue, tax neutrality (economic efficiency)
and, under certain circumstances and to a limited extent, vertical equity.
The base of existing taxes should be broadened at the same time that tax
administration reform is carried out. Base broadening is compatible with a number
of economic objectives. It can increase revenues and improve the simplicity,
neutrality and equity of the tax system.
The use of the tax system for special tax preferences should be carefully evaluated.
Using the system to provide tax incentives (tax expenditures) usually causes a
serious drain on the national treasury by conferring windfall gains on existing
activities or by shifting resources to tax-preferred activities.
Tax reform must take into account the initial conditions at home and abroad. In
reforming their tax systems, developing countries are severely constrained not only
by their own institutional settings but also by the tax structure in capital-importing
countries. Moreover, the circumstances in many developing countries are usually
such that they would experience serious transitional difficulties if the tax system
were to be redesigned from scratch. Developing countries must therefore take into
account initial conditions at home and abroad.
The credibility of the tax regime is the key to the success of any tax reform. A
stable tax policy environment encourages businesses to take a longer-term
perspective in their finance and investment decisions. Making tax changes without
giving adequate consideration to transitional arrangements can undermine the
University of Ghana http://ugspace.ug.edu.gh
62
credibility of the tax regime. Therefore, transitional arrangements require much
more careful analyses than they have hitherto been given in developing countries.
In addition, tax changes must be presented as part of longer-term strategy to
improve the public sector environment for the private sector. The tax regime will
gain the confidence of business if more attention is paid to the preparation and
analysis of reforms, advance consultation, providing a reasonable period of
adjustment prior to implementation and ensuring consistency of the reform
measures.
The tax reform process must be well coordinated. Coordinated tax reform offers
significant advantage over isolated piecemeal tinkering with the tax system. A
coordinated reform ensures that individual tax changes will be consistent with the
central objective. For example, reduction in tariffs without a corresponding
increase in other taxes, generally of a value added type, can increase the fiscal
deficit and exacerbate macroeconomic difficulties. Furthermore, to improve
economic performance in general, tax reform should be closely integrated with
structural adjustment measures.
More often, policy advice directed towards countries desiring to reform their tax systems
has emphasized the introduction of either new taxes or new rates on existing bases, more
stringent administrative changes to seal loopholes that encourage tax evasion, and the need
to widen tax bases and reduce exemptions. Other measures include the allocation of more
budgetary resources to the revenue authorities, paying higher salaries to revenue officers
and imposing relatively prohibitive penalties. Generally, special attention is also directed
University of Ghana http://ugspace.ug.edu.gh
63
to the simultaneous reduction of the tax rates and widening of the base as one of the most
effective approaches for addressing the twin problems of the disincentive effects of high
marginal tax rates and tax evasion.
Tax reform measures are mainly undertaken in order to restore buoyancy to revenues,
strengthen modern taxes, and drastically reduce the complexity and lack of transparency of
the system (World Bank, 1990). The main factors contributing to an improved revenue
performance are changes in tax legislation, tax administration and minimal tax evasion
(Morrisset and Izquierdo, 1993).
3.6 Conclusion
The most important way to ensure an automatic increase in the ratio of government
revenues to gross domestic product is to have a tax structure such that the marginal ratio
exceeds the average ratio or in other words an income-elasticity greater than unity. The
primary interest of those concerned with domestic resource mobilization must therefore be
in the behaviour of the tax system over time.
University of Ghana http://ugspace.ug.edu.gh
64
CHAPTER FOUR
METHODOLOGY AND DATA USED IN THE STUDY
4.0 Introduction
This chapter comprises two sections. The first outlines the theoretical background and the
analytical framework upon which the empirical study is built and discusses application of
the dummy variable method which is used to determine the elasticity of the tax system in
this study. The second looks at the estimation techniques and empirical results from the
data analysis. These entail sources of data, a description of variables of interest and their
proxies. It also presents the characteristics of the time series data.
4.1 Analytical Framework
As governments embrace tax reforms and budget rationalization programmes, it is
necessary that revenue structures designed are flexible enough to guarantee increased
revenue during the growth process without necessarily resorting to discretionary policy or
inflationary financing. Such structures would ensure that tax revenues grow faster that
national income.
Theoretically, predominant taxes in revenue must be those which are highly elastic with
respect to national income. This feature when present in a tax system can be applied to
mitigate the dangers of perpetual fiscal imbalances. High revenue productivity is usually
considered as one of the criteria of a good tax system.
University of Ghana http://ugspace.ug.edu.gh
65
To be able to study the revenue mobilization performance of the mining sector fiscal
regimes, the concepts of tax buoyancy and tax elasticity will be employed as in Singer
(1965) and Mansfield (1972).
4.2 Tax Elasticity
According to Goode (1993), elasticity of a tax refers to the relationship between the
proportional changes in tax revenue and a broad measure of national income or output. It
refers to changes in the tax revenue that occur automatically with no changes in tax
administration, tax rate or exemptions. Tax elasticity tries to reconstruct what would have
happened if there had been no changes in the tax rules. In other words, what tax revenue
would have been if last year‘s laws continued to apply this year? Mathematically, it can be
expressed as
WhereT is tax revenue from mining sector, Y is mining sector GDP and Ety is income
elasticity of tax.
The main use of tax elasticities is to identify which taxes are naturally elastic – i.e., which
taxes will yield more revenue as income (GDP) rises, even if rates are not changed
annually. Elastic taxes are generally considered to be desirable, because they reduce the
need to tinker with the tax system every year. Tax elasticities are unit-free, and so may be
compared across countries without further modification.
University of Ghana http://ugspace.ug.edu.gh
66
Tax elasticities are calculated for total tax revenue and can equally be constructed for
individual taxes with some degree of success. The procedure is first to generate a revenue
series, assuming no change in the tax rates from year to year, and then one may apply
methods to compute the elasticity.
An elastic tax system is one where the incremental tax revenue to national income ratio is
greater than the average tax to national income ratio given that the tax structure does not
change. A given tax is said to be elastic if the marginal tax rate exceeds the average tax
rate, given that the average share of tax base in national income remains constant. The tax
elasticity will be unitary if the marginal and the average rates of taxation are equal and it
becomes inelastic if the marginal is less than the average rate of taxation.
Conventionally, the elasticity of total tax revenue to income is presented in an aggregate
form as a single value, but realistically the overall elasticity of a tax system is the weighted
average of the sum of individual tax elasticity that responds in various ways to changes in
income. This implies that an evaluation of the overall tax elasticity must begin with an
examination of the individual elasticities. Mansfield (1972) defined these elasticities as
follows:
Elasticity of total tax revenue to income:
Elasticity of kth
individual tax to income: (1)
University of Ghana http://ugspace.ug.edu.gh
67
Elasticity of the kth
individual tax to base:
(2)
Elasticity of the kth
individual base to income:
(3)
Where Tt is total tax revenue, Y is income, Bk is base of kth
tax, Tk is revenue from kth
tax.
(4)
Where subscripts 1, 2 and n refer to the different individual taxes which are expressed as a
ratio of total tax revenue indicated by the subscript t to give the individual tax weight.
The elasticity of any individual tax with respect to income is decomposed into the product
of the elasticity of the tax-to-base and the elasticity of the base-to-income.
Mathematically,
(5)
Weighting each individual tax for its share in total tax and summing for all the individual
taxes, we obtain:
University of Ghana http://ugspace.ug.edu.gh
68
(6)
This equation implies that the elasticity of total revenue in a system of n taxes partly
depends on the product of the elasticity of tax base to income for each individual tax,
weighted by the importance of each tax in the overall tax system (Osoro, 1995).
Tax elasticity excludes the impact of discretionary policy changes. The most important use
of tax elasticity is to identify which taxes are naturally elastic. An elastic tax system
enables the public sector to appropriate a growing share of the marginal increases in
income. With an inelastic tax system, a rising public expenditure may be financed either
through higher money supply with the attendant problems of inflation and balance of
payment crises, or by annual upward revision of the existing tax rates. Also an elastic tax
system is considered as efficient and acts as an instrument of stabilization.
4.3 Buoyancy of Tax
The buoyancy of a tax system reflects the total response of tax revenue to changes in tax
policies over time (Osoro, 1995). The buoyancy can be estimated by regressing the tax
revenue on the tax base such as national income (GDP) after taking the natural logarithm
for each of them. It measures the relationship between the proportional changes in tax
revenue and those of the tax base and therefore, gives a quantitative measure of the
effectiveness of tax policy in terms of stimulating public resources (Teera, 2002).
University of Ghana http://ugspace.ug.edu.gh
69
The global buoyancy of a tax system is usually measured by the proportional change in
total tax revenue with respect to the proportional change in national income and can be
expressed as;
Where T is total tax revenue, Y is income or GDP. The global buoyancy can be
decomposed into individual tax buoyancy as;
Where Tt = T1 + T2+ …. + Tn and n is the number of taxes. The global buoyancy is a
weighted sum of the individual tax buoyancy (Sohato, 1961) and used to obtain the
elasticity of tax revenue with respect to tax-to-base and base-to-income as:
And
Thus, global buoyancy
Measurement of tax elasticity is similar to buoyancy of tax with the difference being that,
when no attempt is made to control for discretionary measures that alter the tax rate and/or
University of Ghana http://ugspace.ug.edu.gh
70
base, then the responsiveness of tax revenue to changes in income is the buoyancy and
controlling for such measures give estimates of tax elasticity. It follows that a buoyant
(elastic) tax is the one whose buoyancy (elasticity) is greater than one. According to
Twerefou, Fumey, OseiAssibey and Asmah (2010), discretionary tax measures (DTMs)
are under the manipulation of the policy maker and are generally due to changes in tax
rate, base definition as well as changes in collection and enforcement of tax law, while
non-discretionary changes are due to the natural growth of the economy. When the
elasticity of major revenue sources remains low despite tax reforms either due to low base,
or due to evasion or avoidance, governments raise additional resources through
discretionary measures. Then the growth of tax revenues comes through high buoyancy
rather than through elasticity.
4.4 Model Specification
Two methods have traditionally been employed to estimate tax elasticities. These are the
historical time-series tax data (HTSTD) adjusted to discretionary tax measures (DTMs)
and the unadjusted HTSTD with time trends or dummy variables as proxies for DTMs.
The usual practice in the former technique is to run the proportional adjustment for
cleansing the HTSTD as in Mansfield (1972) and Kusi (1998) or to use the constant rate
structure as in Andersen (1973) and Choudhry (1979). In the latter technique, a Divisia
index has been used as in Choudhry (1979) and sometimes simple or mixed dummies are
used as proxies for each DTM over the estimation period as in Singer (1968), Artus (1974)
and Twerefou et al (2010).
University of Ghana http://ugspace.ug.edu.gh
71
This study adopts the unadjusted HTSTD with dummy variables incorporated as proxies
for discretionary tax measures as in Singer (1965) to evaluate the buoyancy and elasticity
of the tax system due to its non-intensive data requirement and the non-requirement of
disaggregated data.
By specifying a Cobb-Douglas function of the form;
(7)
OLS is then applied to estimate the parameters of α and β, where the coefficient β is an
estimate of tax buoyancy and ∈ is the stochastic term. From the above relationship,
elasticity estimates are obtained by replacing historical tax revenue series by tax revenue
series at a constant tax structure. The revenue at a constant tax structure is obtained from
the historical series of tax revenue by cleaning the tax series for effect of changes in tax
rates and tax base during the reference period. This requires information on the revenue
effects of DTMs introduced in the reference period which is under the manipulation of
policy maker unlike non-discretionary changes that are due to natural growth of the
economy.
Thus, tax buoyancy/elasticity can be decomposed into two:
(8)
University of Ghana http://ugspace.ug.edu.gh
72
(9)
Where TRk is the unadjusted HTSTD of the kth
tax, Bk is tax base for the kth
tax, Y is
nominal mining sector GDP which is the overall base, ak is elasticity of the kth
tax to its
base, δk is the elasticity of kth
tax base to income, a0 and δ are constants and and µ are
stochastic error terms.
With the help of the two kinds of elasticity we obtain the elasticity of the kth
tax to income
ΗK as the product of αkandδk. To capture the effects of the reforms in the short run, a
dummy variable D is introduced into the decomposed elasticity equations above. The
modified short run elasticity and buoyancy functions are:
(10)
(11)
Where and are the dummy coefficients and the summation sign represents the total
DTMs over the period under study.
University of Ghana http://ugspace.ug.edu.gh
73
4.5 The Dummy Variable Approach
The Dummy Variable Approach makes use of unadjusted historical time series tax data
with time trends or dummy variables incorporated as proxies for DTMs. This is done to
capture the elasticities. From equation (7), the empirical model can be shown as:
(12)
Where TRtk denotes tax revenue for k
thtax, β is the elasticity and D is the dummy variable,
which takes a value of one for discretionary tax measures and zero otherwise.The
summation accounts for the multiple discretionary tax changes over the sample period.
Two dummies are considered in this study – a tax reform dummy, D1996 introduced to
capture fiscal reforms undertaken in 1996 and an interactive term Dslopewhich is the slope
of the tax revenue function as a result of the tax reforms. Dslope is defined as TR×D1996
which ensures the linearity of the model. The lagged base is also introduced to account for
the administrative efficiency, or otherwise, in tax collection as tax policy guidelines
announced in budget speeches normally require time to be implemented. If there are severe
administrative lags, the lagged value will be more significantly related to the dependent
variable.
The Engle-Granger two steps cointegration procedure was used to establish the long-run
relationship between the relevant variables and to generate the error correction term for the
aggregate and individual tax functions. The Breusch-Godfrey Serial Correlation LM test is
preferred over the Durbin Watson Statistic in detecting the presence of serial correlation in
the data set.
University of Ghana http://ugspace.ug.edu.gh
74
4.6 Estimation Techniques
This section is on data source and various tests of time series of the data (i.e., unit root and
cointegration tests).
4.6.1 Taxes Considered
The approach of analysing the mining tax system in terms of buoyancy and elasticity of
this study begins with a look at the aggregate tax revenue function and the individual taxes
that form the total tax. A brief definition of the total tax revenue and its constituent tax
units are presented below:
Total Tax Revenue (TTR): This is the summation of Pay-As-You-Earn Tax (PYTAX),
Company Tax (COTAX) and Royalty Tax (ROYTAX) of the mining sector in Ghana.
That is, TTR = PYTAX + COTAX + ROYTAX
4.6.2 Choice of tax bases and Data Sources
The different types of taxes to be considered include the aggregate tax revenue function
and the individual taxes that form the total tax in the analysis of buoyancy. With regards to
the proxy tax base for the taxes considered in this study, there is no precise legal definition
for tax bases and this is a limitation to the tax data set. It is argued that proxy bases
constitute a reasonable representative of the component of national wealth of transactions
upon which a particular tax can be assessed (Mtatifikolo, 1990).
University of Ghana http://ugspace.ug.edu.gh
75
Secondary Data on all tax revenues were sourced from Ghana Revenue Authority and the
Minerals Commission of Ghana for the period under study. Quarterly Digest of Statistics
and Economic Survey Reports were sourced from the Ghana Statistical Services. These
publications contain annual time series data on economic classification of central
government revenue among others.
The IMF source of data was derived from the Government Finance Statistics Yearbook
which contains time series data on central government operations of member countries.
Data from this source was used to supplement those of GoG sources and are considered
adequate for estimating tax buoyancy and elasticity of the Ghanaian tax system.
Table 4.0: Description of Various Taxes and Their Proxy Bases
Tax Type Definition Proxy Base Definition Source
Total Tax
Revenue
(TTR)
Summation of all
individual tax
revenues
contributions by the
mining sector
Mining sector
GDP
Share of GDP
attributable to
the Mining
sector
Quarterly
Digest of
Statistics
Personal
Income Tax
(PYTAX)
Tax levied on
personal income and
earnings of
employees based on
the principle of Pay
As You Earn
(PAYE)
Mining Share of
Current
Personal
Income
(CUPY)
Private
employee
emoluments in
the mining
sector
Quarterly
Digest of
Statistics &
Economic
Survey Reports
Corporate
Income Tax
(COTAX)
Tax on earnings of
mining companies
Mineral Exports
(MEXP)
Value of
exports of all
gold, diamond,
manganese and
bauxite
IMF & Bank of
Ghana Annual
Reports
Royalties
(ROYTAX)
Production base tax
by mining lease
holders to the
government through
GRA
Mineral
Exports(MEXP)
Value of
exports of all
gold, diamond,
manganese and
bauxite
IMF & Bank of
Ghana Annual
Reports
University of Ghana http://ugspace.ug.edu.gh
76
4.7 Data Characteristics
The time series data for the study is analysed to ensure its stability and long run
relationship between the variables as well as short run disequilibrium. These are performed
in line with the following steps:
4.7.1 Unit Root Test for Stationarity
Non-stationarity of a time series data is regarded as a problem in an empirical analysis, so
as a short-hand rule, checking if the regression will give a reliable result is to ensure that
the data is stationary. The variables may be non stationary at level, in this case a retest at
first difference is carried out and if still not stationary, a second difference test of
stationarity is performed and so on; the number of times a data is differenced gives the
order of integration. The unit root test is used to test for the stationarity of the data and this
begins by specifying an Augmented Dickey-Fuller (ADF) test equation as follows:
Where Xt is any of the variables used in the model, a0 is constant, t is the linear trend, and
the lag of ΔXt incorporated in the model forms the augmentation. The i is the optimal lag
length that is set to ensure that any autocorrelation in ΔXt is absorbed and the error term is
assumed distributed as white noise (Dickey and Fuller, 1981).
University of Ghana http://ugspace.ug.edu.gh
77
A regression test is run on each variable against its lagged ones and the lag difference
terms of the same. A line graph of each variable is plotted to observe their characteristics
before choosing the constant term, the constant term and linear trend t, or none in the
equation. The appropriate test with respect to the data characteristics is applied as follows:
if the calculated t-ratio of the coefficient β with negative sign is less than its critical value
then Xt is said to be stationary.
4.7.2 Cointegration analysis
Non-stationary variables which are not cointegrated lead to nonsensical result with no
meaningful economic interpretation, this is called spurious regression (Granger and
Newbold, 1974) and this means not all variables share trend can be explained
meaningfully in terms of economics. In order to avoid such spurious regression, the
method used in this study is the Pedroni (Engel-Granger based), 2004 technique. The
existence of a cointegrating relationship implies that the regression of the non-stationarity
series in their level yields meaningful and not spurious results. However for cointegration
to exist, the non-stationarity series must be integrated of the same order. The Engle-
Granger two-step cointegration procedure is adopted to test for cointegration in this study.
This method involves an adjustment process that prevents errors from becoming
indefinitely in the long run relationship between the dependent variables (tax variables)
and the independent variable (tax bases). In other words, when non-stationarity variables
move together, in the long run there exists a stationary linear combination of these
variables. The first step in this test is to estimate cointegrating static (long run) models
University of Ghana http://ugspace.ug.edu.gh
78
using the OLS method. The second step involves generation of residuals from the static
models which are then individually evaluated in terms of their order of integration using
the DF or ADF unit root test. It should be noted however that the usual ADF critical values
are inappropriate (Gujarati, 2003), so Engel and Granger (1987) came up with the
appropriate critical values against the ADF t-statistic can be resolved (Charemza and
Deadman, 1997).
4.7.3 Error Correction Model (ECM)
To determine the short run elasticity and buoyancy coefficients of the respective
cointegrating regression equation, an over-parameterized ECM version of the relevant
equations is used. The error correction terms (ECTs) is derived, as the lagged residuals
value generated from the estimated long-run cointegrating equations. A ‗general to
specific‘ procedure of parsimonious model specification is followed. The variables in the
over-parameterized tax elasticity and buoyancy models with the exception of ECTs and the
dummy variable is set at two lag to economize on the degrees of freedom. The Schwarz
Information Criterion (SIC) is used to drop statistically insignificant variables from the
models in order to come up with more preferred specifications.
University of Ghana http://ugspace.ug.edu.gh
79
5
6
7
8
9
90 92 94 96 98 00 02 04 06 08 10
TTR
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
90 92 94 96 98 00 02 04 06 08 10
PYTAX
4
5
6
7
8
9
90 92 94 96 98 00 02 04 06 08 10
COTAX
5.2
5.6
6.0
6.4
6.8
7.2
7.6
8.0
8.4
90 92 94 96 98 00 02 04 06 08 10
ROYTAX
8.6
8.7
8.8
8.9
9.0
9.1
9.2
90 92 94 96 98 00 02 04 06 08 10
GDP
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
90 92 94 96 98 00 02 04 06 08 10
MEXP
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
90 92 94 96 98 00 02 04 06 08 10
CUPY
CHAPTER FIVE
PRESENTATION AND DISCUSSION OF RESULTS
5.0 Introduction
In this chapter, a summary of empirical results from econometric analysis of ordinary least
square regression estimates of the tax buoyancy and income elasticity models of the
Ghanaian mining sector tax system using time series are presented. The analyses were
done using Eviews 7 econometrics software.
5.1 Results of Unit Root Test
Unit root test results for the ADF in both levels and first differences of the variables are
reported in Table 5.0 below and they are used to examine the stationarity of the series. The
graphical representations of all the variables are reported.
Figure 5.0: Graphical Representation of Log Variables in Levels
University of Ghana http://ugspace.ug.edu.gh
80
Table 5.0: Results of Augmented Dickey-Fuller (ADF) Test for Unit Root
Variable Definition of
Variable
ADF
Statistic
1% Critical
Value
5%
Critical
Value
Order of
Integration
LTTR Total Tax Revenue -1.8702 -4.4983 -3.6584 I(1)
DLTTR -4.2404** -4.5326 -3.6736 I(0)
LPYTAX Personal Income
Tax
-2.7888 -4.6162 -3.7105 I(1)
DLPYTAX -4.2525** -4.6679 -3.7332 I(0)
LCOTAX Company Tax 0.0591 -3.8085 -3.0207 I(1)
DLCOTAX -4.6665** -3.8868 -3.0522 I(0)
LROYTAX Royalty Tax -2.3519 -4.4983 -3.6584 I(1)
DLROYTAX -5.0098** -4.5326 -3.6736 I(0)
LGDP Gross Domestic
Product
-2.0499 -3.8085 -3.0207 I(1)
DLGDP -5.0242** -4.5326 -3.6736 I(0)
LMEXP Mineral Exports -2.7870 -4.5326 -3.6736 I(1)
DLMEXP -3.7243* -4.5326 -3.6736 I(0)
LCUPY Current Personal
Income
-3.1023 -3.5326 -3.6736 I(1)
DLCUPY -3.7320* -4.5326 -3.6736 I(0)
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
Unit root test results for the ADF in both levels and first differences of the variables in
Table 5.0 above showed that total tax revenue, personal income tax revenue, company tax,
royalty tax, GDP, mineral exports and current personal income were integrated of order
zero implying that there was no presence of unit roots in first differenced form of the
variables. This also implies that stationarity of the variables was attained in first
University of Ghana http://ugspace.ug.edu.gh
81
difference. All variables except Mineral Exports and Current Personal Income were
significant at 1%. The two were subsequently significant at 5%.
5.1.1 Results of Cointegration Test
The Engle-Granger Two-Step was applied to test for the existence of cointegration. The
procedure was applied on the residuals after obtaining the static long-run regression for
each stochastic equation. The residuals are then saved and tested for stationarity. A non-
stationary residual implies that the variables in the equation generating the residual are not
cointegrated and the results are presented in Table 5.1 below.
Table 5.1: Engle-Granger Two Step Cointegration Test: Buoyancy Results
Tax Equation Residuals ADF-
Statistic
1% critical
value
5% critical
value
Inference
TTR TTRres -3.8378* -4.5716 -3.6908 I(0)
PYTAX PYTAXres -4.6710* -4.7284 -3.7597 I(0)
COTAX COTAXres -3.9434** -3.8085 -3.0207 I(0)
ROYTAX ROYTAXres -3.4930* -3.8085 -3.0207 I(0)
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
University of Ghana http://ugspace.ug.edu.gh
82
-1.2
-0.8
-0.4
0.0
0.4
0.8
90 92 94 96 98 00 02 04 06 08 10
TTRRES
-1.0
-0.5
0.0
0.5
1.0
90 92 94 96 98 00 02 04 06 08 10
PYTAXRES
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
90 92 94 96 98 00 02 04 06 08 10
COTAXRES
-.8
-.4
.0
.4
.8
90 92 94 96 98 00 02 04 06 08 10
ROYTAXRES
Figure 5.1: Line Graphs for the Buoyancy Equation
University of Ghana http://ugspace.ug.edu.gh
83
From the Table 5.1above, total tax residuals are stationary at level suggesting the presence
of a cointegrating relationship between total tax revenue and mining sector GDP.
Similarly, all other tax residuals attained stationarity at level. ADF statistics for all tax
residuals were either significant at 1% or 5%. All except company tax residuals were
significant at 5%. Conclusively, cointegration test for the buoyancy equation (Table 8)
showed that all residuals were stationary in levels which support the existence of a
cointegrating relationship in the buoyancy estimation equations of all the variables.
Table 5.2: Engle-Granger Two Step Cointegration Test: Tax-to-Base Elasticity
Results
Tax Equation Residuals ADF-
Statistic
1% critical
value
5% critical
value
Inference
TTR TTRres -3.8378* -4.5716 -3.6908 I(0)
PYTAX PYTAXres -3.7852* -3.9204 -3.0656 I(0)
COTAX COTAXres -3.5943* -3.8867 -3.0522 I(0)
ROYTAX ROYTAXres -4.1706** -3.8574 -3.0404 I(0)
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
University of Ghana http://ugspace.ug.edu.gh
84
-1.2
-0.8
-0.4
0.0
0.4
0.8
90 92 94 96 98 00 02 04 06 08 10
TTRRES
-.8
-.6
-.4
-.2
.0
.2
.4
.6
90 92 94 96 98 00 02 04 06 08 10
PYTAXRESTB
-2
-1
0
1
2
90 92 94 96 98 00 02 04 06 08 10
COTAXRESTB
-.8
-.6
-.4
-.2
.0
.2
.4
90 92 94 96 98 00 02 04 06 08 10
ROYTAXRESTB
Figure 5.2: LineGraphs for the Tax-to-Base Residuals
From Table 5.2above, all total tax residuals were stationary at level suggesting the
presence of a cointegrating relationship between total tax revenue and their respective
proxy bases. This showed that Pay-As-You-Earn tax had a cointegrating relationship with
current personal income (CUPY). Company and royalty taxes also shared a cointegrating
relationship with mineral exports. All other tax residuals attained stationarity at first
difference. ADF statistics for all tax residuals were either significant at 1% or 5%. All
except royalty tax residuals were significant at 5%. Conclusively, cointegration test for
the tax-to-base elasticity equation (Table 5.2) showed that all residuals were stationary in
levels which supported the existence of a cointegrating relationship in the tax-to-base
buoyancy estimation equations of all the variables.
University of Ghana http://ugspace.ug.edu.gh
85
Table 5.3: Engle-Granger Two Step Cointegration Test: Base-to-Income Elasticity
Results
Tax
Equation
Residuals ADF-
Statistic
1% critical
value
5% critical
value
Inference
CUPY CUPYres -3.8218* -3.8868 -3.0522 I(0)
MEXP MEXPres -5.7857** -3.8574 -3.0404 I(0)
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
From Table 5.3 above, current personal income and mineral exports share of a
cointegrating relationship with GDP as their residuals are both integrated of order zero.
Current personal income residuals represented by their ADF statistic were significant at
5%. Mineral export residuals were however significant at 1%. Thus, the cointegration test
for the base-to-income elasticity equation (Table 5.3) showed that all residuals were
stationary in levels which supported the existence of a cointegrating relationship in the tax-
to-base buoyancy estimation equations of all the variables.
5.2 Regression Results
Below are the estimated regression models for the long and short run buoyancy and
elasticity models. Diagnostic tests showed that residuals are normally distributed and
stable. Where the probability value of the Breusch-Godfrey serial correlation LM test
suggests high degree of residual autocorrelation, the problem was corrected by the use of
University of Ghana http://ugspace.ug.edu.gh
86
Generalised Least Squares estimation procedure assuming an AR(1) process. Tables
showing the Breusch-Godfrey LM test are shown in the appendix.
The study shows that variables were cointegrated; that is, there is a long term relationship
between them. This also means that in the short run there may be disequilibrium.
Therefore the error correction term is added as an explanatory variable.
For example since COTAX and MEXP are cointegrated, the short run equation is
expressed as
5.2.1 Tax Buoyancy: Long Run and Short Run
Table 5.4 below shows results of the long run and short run regressions of the estimated
tax buoyancy of the mining sector tax system for the period 1990 to 2010. The estimated
regression models performed well in terms of goodness of fit and joint significance with
Adjusted R2 of above 60% and F-statistic being significant at 1% respectively.
University of Ghana http://ugspace.ug.edu.gh
87
Table 5.4: Estimated Tax Buoyancy Results of the Tax system in Ghana’s Mining
Sector
Short Run Dynamics Long Run Dynamics
Tax Variables Coefficient
(Buoyancy)
DW Ṝ 2 Variables Buoyancy DW Ṝ 2
TTR
LGDP-2** 0.69 2.31 0.92 LGDP** 0.78 1.59 0.98
ECMTTR-
1**
-0.13
PYTAX
LGDP-1** 0.22 2.57 0.61 LGDP** 1.03 2.12 0.97
ECMPYT-
1**
-0.31
COTAX
LGDP-1** 1.66 2.09 0.74 LGDP* 2.14 1.98 0.90
ECMCOT-
1**
-0.34
ROYTAX
LGDP-1** -1.81 2.43 0.72 LGDP** 0.55 2.30 0.98
ECMROY-
1**
-0.66
Source: Authors’ Computation. Note: * denotes significance at 5% and ** denotes significance at 1%.
Results from the long run estimation showed that all but royalty taxes were buoyant.
Overall, total tax revenue showed a buoyancy of 0.78 which implies that as GDP increases
by 100 percent; tax revenue rises by 78%. This illustrates that tax revenue from the taxes
grow less faster than the GDP and could imply that the discretionary changes implemented
have generally been ineffective at increasing tax revenue for majority of the taxes
compared to proportionate change in GDP. This result is in line with a study done by
Chipeta (1998) which established the overall buoyancy of Malawi to be 0.95 (less than
unity). This result contradicts a study by Gillani (1986) which found Pakistan‘s tax
system‘s long run buoyancy to be elastic (1.17). The finding of 0.78 can be attributed to
the low buoyancies of royalty taxes which are a key component of the total mining tax
University of Ghana http://ugspace.ug.edu.gh
88
system. Pay-As-You-Earn tax however had a positive and significant coefficient of 1.03.
This indicates that PAYE tax revenue grows faster than GDP in the long run. Again, this
result contradicts Gillani (1986) study which attained a buoyancy of 0.82 for income tax in
Pakistan but confirms the work of Bilquees (2004) which found a long run buoyancy of
1.15 for the non-agricultural sector of Pakistan. Buoyancy results for company tax revenue
of 2.14 showed that company tax revenue grows more than double the growth rate in GDP.
This result can be attributed to the fact that the degree to which underestimation of profits
by companies with respect to company tax is curbed through external auditing. Tax audits
conducted in the mining companies place emphasis on whether profits declared are
underestimated or otherwise. This result is supported by similar findings by Kusi (1993)
which found a post-tax reform company tax buoyancy of 1.36, Gillani (1986) which found
corporation tax buoyancy of 2.25. Similarly, the work of Twerefouet al. (2010) which
found a long run buoyancy of 1.20 for corporate tax is confirmed by the result shown in
Table 11. Royalty taxes returned a long run buoyancy of 0.55 which implies that a hundred
percentage change in GDP causes revenue sourced from royalties to change by 55%.
The negative and significant coefficients of the error correction term lagged one period
(ECT-1) for all models could imply that there are economic forces that operate to restore
the long-run equilibrium path of revenue following short-run disturbances. For example,
Total Tax Revenue (TTR) and Company Tax (COTAX) regression models coefficients of
-0.13 and -0.34 respectively means that Total Tax Revenue model adjusts to its long run
equilibrium at a speed of 13% whiles Company Tax model does same at 34%. Similarly,
PYTAX and ROYTAX regression model coefficients of -0.31 and -0.66 means that the
respective tax revenue models adjust to their long run equilibrium at a speed of 31% and
University of Ghana http://ugspace.ug.edu.gh
89
66%. The results showed that in the short run, GDP in its first and second lags have
positive influence on TTR, PYTAX, COTAX and negative influence on ROYTAX. Short
run buoyancy for the overall tax system and individual tax categories were less than unity
except company tax (COTAX). Company tax being very buoyant in the short run suggests
that this tax revenue grows faster than GDP and this could imply that the fiscal changes in
company tax rates have been effective at increasing the tax revenue compared to
proportionate change in output. Pay-As-You Earn tax (PYTAX) and Royalty tax recorded
short run estimates of 0.22 and -1.81 respectively. The results for COTAX, PYTAX and
TTR are in line with Kusi (1993) buoyancy estimation for Ghana and Muriithiet al. (2003)
buoyancy estimation for the Kenyan tax system.
5.2.2 Decomposed Buoyancy into Tax-to-Base and Base-to-Income Elasticities
Table 12 below gives the Tax-to-Base results for the long run static and short run dynamic
regressions for the Ghanaian mining tax system over the sample period 1990-2010. The
estimated Tax-to-Base elasticity models performed well in terms of goodness of fit and
overall significance with Adjusted R2 of above 70% and the F-statistic significant at 1%
respectively. Where the Breusch-Godfrey LM serial correlation test suggests
autocorrelation, the problem was dealt with by using Generalised Least Squares estimation
procedure assuming an AR(1) process.
University of Ghana http://ugspace.ug.edu.gh
90
Table 5.5: Estimated Tax-to-Base Elasticity of the Mining Sector Tax System
Short Run Dynamic Long Run Dynamic
Tax Variables Coefficient
(Elasticity)
DW Ṝ 2 Variables Elasticity DW Ṝ 2
TTR
LGDP-2** 0.69 2.31 0.92 LGDP 0.78 1.58 0.97
ECMTTR-
1**
-0.13
PYTAX
LCUPY-
1**
0.05 1.75 0.85 LCUPY 0.31 2.27 0.97
DUM** 0.48
COTAX
LMEXP-
1**
0.30 2.07 0.71 LMEXP 0.33 1.88 0.93
ECMCOT-
1**
-0.56
ROYTAX
LMEXP-
1**
0.25 1.61 0.81 LMEXP 0.26 2.61 0.99
DUM** 1.24
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
It is observed from the Table 5.5 above that, in the long run, the overall tax system and the
various tax categories were inelastic as indicated by the tax-to-base elasticity coefficients
which were less than unity meaning that a percentage change in their respective bases
resulted in less than a percentage change in tax revenue from the various taxes. Pay-As-
You-Earn tax had a tax-to-base elasticity of 0.31 which suggests that current personal
income grows faster than the PYTAX. Given a percent change in current personal income,
PYTAX changes by 0.31 percentage points. Similarly, COTAX and ROYTAX tax-to-base
coefficients of 0.33 and 0.26 were inelastic with respect to Mineral exports. Thus mineral
export growth exceeds the tax revenue from these too sources. These results are in
University of Ghana http://ugspace.ug.edu.gh
91
consonance with Chen (2008) which found out the Chinese tax system yielded an inelastic
tax-to-base elasticity in the long run. However, Twerefouet al. (2010) found the opposite
to be the case in the overall Ghanaian tax system. This is largely due to choice of tax bases
as well as the different nature of the mining sector from the overall economy where for
example no import taxes in the mining sector. In addition, higher cost of production and
general, selling and administrative expenses stemming from the cedi depreciation against
the US Dollar (especially from 1998 to 2009) resulted in lesser profits and hence lesser
corporate tax even though revenue from mineral exports increased.
The short run results showed that the individual tax categories are not elastic to their bases
as reflected by the coefficients which are less than unity. GDP in its second lag exerts an
elasticity of 0.69 with respect to total tax revenue. Total tax revenue also adjusts to its long
run equilibrium at a speed of 13% as indicated by the error correction term. Similarly,
COTAX had a less than unit elasticity of 0.30 and attained its long run equilibria at a speed
of 56%. The results also showed that Company tax was however elastic to its base in the
short run. The one period lagged mineral export‘s positive influence of 0.30 for the
COTAX could be a reflection of arrangements whereby tax returns are submitted in
arrears. The dummy variable coefficient of 0.48 and 1.24was also significant for PYTAX
and ROYTAX and this indicates that there is a short run impact of the fiscal regime on
these taxes in terms of improving tax-to-base elasticity. The error correction term
coefficients were also statistically significant for all total tax and company tax variables.
The inelastic nature of tax-to-base relationship of this study is confirmed by Kargboet al.
(2012) study of the tax elasticity in Sierra Leone which found out that total tax system-to-
base elasticity was 0.81 in the short run.
University of Ghana http://ugspace.ug.edu.gh
92
Table 5.6: Estimated Base-to-Income Elasticity of the Mining Sector Tax System
Short Run Dynamic Long Run Dynamic
Base Variables Coefficient
(Elasticity)
DW Ṝ 2 Variables Elasticity DW Ṝ 2
CUPY
LGDP-1 0.48 2.15 0.47 LGDP 1.00 1.65 0.99
ECMCUPY-
1
-0.04
MEXP
LGDP-2 0.20 1.73 0.57 LGDP 0.94 1.33 0.99
ECMMEXP-
1
-0.48
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%.
Long run decomposition of buoyancy into base to income elasticity from the Table 5.6
above shows that base to income buoyancy is unitary elastic (elasticity of 1) for current
personal income and inelastic (elasticity of 0.94) for mineral exports. In the short run,
GDP in its first and second lags have positive influences on the current personal income
and mineral exports respectively. The negative and significant one period error correction
terms also suggests that there is a short run impact of the various fiscal regimes on these
taxes in terms of improving tax-to-base elasticity. The short run model also has all
individual tax categories being base-to-income inelastic as shown by the coefficients being
less than unity. From the results in Table 13 above, short run dynamics will attain long run
equilibria with 4% for current personal income and 48% for mineral exports. These results
are similar to a study by Kusi (1998) which established that base to income elasticity was
inelastic in both the long and short run. It also justifies the concern that the country‘s tax
base in not wide enough to capture tax revenue. According to Twerefouet al. (2010),
similar concerns were raised by Osoro (1993) and Suliman (2005) and the low elasticity
was attributed to the tax base not been wide enough to capture many tax payers due to the
University of Ghana http://ugspace.ug.edu.gh
93
presence of a significant underground economy thereby resulting in excess burden on few
people in the tax net.
5.2.3 Income elasticity of the Mining Tax System in Ghana
Table 14 below shows the results of the long run estimated elasticities of the total tax
revenue and the individual categories of tax using the Mansfield (1972) technique. All
equations have a good fit as indicated by the adjusted R2 and significant probability value
of F-statistics.
Table 5.6: Estimated Tax Elasticity Results of the Tax system in Ghana’s Mining
Sector
Tax Equation Elasticity Reform
Dummy
Slope Dummy Lagged Base Ṝ 2
TTR 0.75**
(0.55)
-0.73** (-
0.15)
0.04** (0.12) 1.24** (0.91) 0.96
PYTAX 0.94* (0.67) -6.58 (-1.20) 0.46 (-1.28) -0.31** (-
0.22)
0.82
COTAX 1.04* (0.70) 9.45* (0.72) -0.70* (-0.82) 2.90** (0.79) 0.71
ROYTAX 0.42 (0.34) -4.28 (-0.95) 0.29 (0.99) 1.07** (0.84) 0.66
Source: Authors’ Computation, Note: * denotes significance at 5% and ** denotes significance at 1%; t-
values are reported in parenthesis.
The long run elasticities of the total tax revenue and the individual categories of tax
indicate elastic and inelastic estimates. The overall mining sector tax elasticity was
inelastic (i.e., 0.75). This illustrates that as GDP increases by a hundred percent, tax
University of Ghana http://ugspace.ug.edu.gh
94
revenue from the mining sector yielded a less than one percent change (i.e., 75%). The
coefficient is significant at 5% level and we can state that the overall tax system is income
inelastic over the long run period. This supports the study by Kargboet al. (2012) which
also found the Sierra Leonean nationwide tax system to be 0.89 thus, inelastic. However,
the overall mining sector tax elasticity with respect to GDP in its first lag, however,
recorded a significant elasticity of 1.24. This could be the result of administrative
efficiencies in the collection of taxes. Reform and slope dummies were significant at 5%
but had negative (-0.73) and positive (0.04) influences respectively. This makes it difficult
to assess the impact of fiscal regime on tax yield since there are opposing influences from
the two dummies.
The elasticity of personal income tax revenue was also inelastic at 0.94. The coefficient is
significant at 1% and illustrates that the progressive tax system implemented in Ghana
changed by 94% for every corresponding 100% change in mining GDP. This can partially
be explained by the fact that income generated from local labour in the mining sector
accounts for a relatively small share of the total value of production mainly because all
mines coming on stream are surface operations that use capital intensive techniques.
Secondly, while the income tax law provides for the taxation of all revenues, whether
national or expatriate, in practice investors have obtained exemption or reduced taxation
on the income of their employed expatriates in their negotiated investment agreements.
This finding is confirmed by the work of Muriithiet al. (2003) which found out an inelastic
coefficient of 0.49 for direct taxes in Kenya. Here also, reform and slope dummies posted
opposing influences of -6.58 and 0.46 respectively making it difficult to assess the impact
of the fiscal changes.
University of Ghana http://ugspace.ug.edu.gh
95
Elasticities for Company tax were found to be greater than unity with a coefficient of 1.04
hence elastic. Alternatively, its elasticity with respect to the GDP in its first lag was
significant at 1% and elastic with a coefficient of 2.90. This implies that a percentage
change in GDP in its first lag resulted in a 2.90 percentage change in company tax
revenue. This is as a result of the regulations in place whereby companies file their
company tax returns in the next year because payment of this tax requires that company
accounts are audited which normally runs into the next year of operations.
Tax elasticity of royalties was found out to be 0.42 for a percentage change in GDP.. It
also had an elastic coefficient with respect to the lagged base of 1.07. Reform and dummy
variables again had opposing influences with a reform dummy coefficient of -4.28 and a
slope dummy coefficient of 0.29.
All taxes possessed positive lagged base and significant at 5% level except for personal
income taxes. This could be due to the result of administrative efficiencies in the collection
of these taxes. The reform dummy was only positively significant for the company tax at
5% level. The insignificant reform dummy coefficients indicate that the changes in fiscal
regime have not impacted on the revenue sources. On the whole, the slope dummies had
opposing influences with reform dummies making it difficult to assess the impact of fiscal
changes on tax yield from the tax revenue sources. The dummy coefficients were not
significant for the PAYE tax and royalty tax. These results are comparable to a study in
Sudan bySuliman (2005) where trade liberalization dummy and slope shift dummy had
opposing influence on income tax yield thereby making it difficult to assess the impact of
liberalization on tax yield from this source.
University of Ghana http://ugspace.ug.edu.gh
96
5.3 Comparison of Tax Buoyancy and Elasticity
Table 5.7: Estimates of Long Run Tax Buoyancy and Elasticity
Tax Buoyancy Elasticity Difference
Total Tax Revenue 0.78 0.75 0.03
Pay-As-You-Earn 1.03 0.94 0.09
Company Tax 2.14 1.04 1.1
Royalty Tax 0.55 0.42 0.13
Source: Author’s Estimation
Table 5.7 above shows the estimated values of buoyancy and elasticity over the study
period. ―Difference‖ was measured as the buoyancy estimates minus elasticity estimates.
The difference was positive for the overall tax system and individual tax categories. This
implies that the various discretionary measures undertaken in the study period have
improved tax yield in these taxes. For example, for a one percent increase in mining sector
GDP, the discretionary tax measures mobilized an additional 0.03 percentage points of
revenue from the overall tax system. For the other taxes, discretionary tax policies yielded
0.09%, 1.1% and 0.13% in PAYE, Company Tax and Royalty Tax respectively for every
growth in mining sector GDP
5.4 Conclusions
A number of findings were obtained from this empirical research. Taxes are not wholly
reactive to changes in income as most elasticity estimates are below unity. The estimates
University of Ghana http://ugspace.ug.edu.gh
97
of buoyancy and elasticity indicate that, overall, the use of discretionary measures has
been relied upon extensively as a source of revenue augmentation in Ghana. The relatively
higher coefficients obtained for the buoyancy estimates compared to the elasticity
estimates point to the role of discretionary measures in maintaining a steady source of tax
revenue for the government. The use of discretionary measures to raise additional tax
revenues is not only peculiar to Ghana as other studies have shown the use of such
measures in less developed economies (Kusi, 1998; Chipeta, 1998; Twerefou et al, 2010;
Muriithi et al, 2003; Osoro, 1993). The tax bases could not expand as mining GDP grows
resulting to low elasticities. Some of the factors that prevented the tax bases to broaden are
numerous exemptions granted to newly established mining companies, tax incentive and
duty waivers.
The short run elasticities being lower than the long run elasticities and the statistical
significance of the error terms showed the speed of adjustment in the tax system. This
trend could explain the presence of administrative lags in the tax collection system as tax
policy guideline announced in budget speeches normally require time to be implemented.
From the regression results, the following are a summary of the findings;
In the short run the tax yield of the mining sector and individual taxes are not
buoyant despite different fiscal regimes except for company taxes which were
buoyant.
In the long run the tax yield of the mining sector was not buoyant but personal
income and company taxes were buoyant. Royalty taxes were also not buoyant in
the long run.
University of Ghana http://ugspace.ug.edu.gh
98
The overall tax yield and individual tax categories were inelastic with the exception
of company taxes.
University of Ghana http://ugspace.ug.edu.gh
99
CHAPTER SIX
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
6.0 Introduction
This chapter focuses on the summary and conclusions of the study. The study implications
and suggested recommendations out of the findings for future fiscal regimes and revenue
mobilization drives that will improve the buoyancy and elasticity of the Ghanaian mining
tax system are also presented in this chapter. Limitation(s) of the study are also discussed.
The Government of Ghana undertook changes in mining fiscal regimes in 1985 and since
then series of other tax initiatives and reforms have taken place, all in a bid to address the
fiscal challenges that have plagued the country and mining sector investment. The study
focused on the tax buoyancy and income elasticity of the tax system in the mining sector
of Ghana as a means of evaluating tax performance and revenue mobilization capacity of
the overall mining tax system and the major mining tax categories with particular
emphasis on the 1996 tax reforms. Time series data over the period 1990 to 2010 was used
and the Engel-Granger method of estimation was employed. The dummy variable
technique was used to control for the discretionary tax measures (DTMs) implemented
over the period.
The study also presented an overview of the tax structure and administration with respect
to the mining sector of Ghana with a summary of fiscal regimes and discretionary
measures undertaken over the sample period as well as some stylized facts about the
University of Ghana http://ugspace.ug.edu.gh
100
mining sector. It was revealed that most reforms were aimed at generating revenue as well
as creating the incentive for private investment in the sector.
Both theoretical and empirical literature on taxation was reviewed. The theoretical
literature emphasized the role of taxation in economic development and also the
importance of buoyancy and elasticity in achieving economic growth. Towards this end
the review focused on some studies that have been done on buoyancy and elasticity and
pointed out the various techniques of evaluating them.
The empirical results and analysis presented in the previous chapter suggests that the fiscal
regimes had different impact on different taxes. Thus, while most of the prominent tax
categories in the system had inelastic yield with respect to the national income in the long
run, company tax had an elastic yield. This implies the proposition that the mining tax
system is not elastic and thus not efficient in mobilising maximum revenue from the
mining sector.
On buoyancy, the mining tax system as a whole did not have a buoyant system in the long
run. Individual tax categories like personal income and company taxes in the long run had
buoyant yield. Royalties, however, did not have a buoyant yield.
University of Ghana http://ugspace.ug.edu.gh
101
6.1 Recommendations
6.1.1 General Recommendations
The empirical results in this study have important implications for the mining sector in
terms of tax performance and revenue yield. To begin with, for the tax system to be
revenue enhancing, the yields of individual taxes should be responsive to changes in
national income as such the major taxes should be highly elastic and buoyant with respect
to national income or proxy bases for any developing country.
The overall tax system was found to be not buoyant and inelastic whiles some of the tax
categories were found to be elastic or buoyant than the others. Subsequently, it calls for
dynamic and strategic policies to target each tax differently. In this case broadening of the
proxy bases of these taxes would help the overall mining tax system and the individual
mining tax categories to be more buoyant and elastic. However, since tax bases are under
the control of the government rather than the tax authorities, measures need to be put in
place by the government to ensure a broad tax base.
Some of these measures may include creating an enabling business environment for
companies to flourish so as to enhance the company tax base, and perhaps making effort at
improving wage emoluments would help boost private income and eventually increasing
the base of personal income tax.
Moreover, it was also revealed that despite the relative successes chalked in the tax system
in terms of revenue mobilization as a result of fiscal regimes, there is still room for further
University of Ghana http://ugspace.ug.edu.gh
102
improvement in the collection; this therefore requires strengthening the capacity of the
revenue agencies to register more eligible tax payers (small scale miners) into the tax net.
To consider the tax buoyancy and income elasticity as decision criteria to ascertain the
performance of the mining sector tax system with regards to revenue generation, it may be
said to have been successful since we cannot lose sight of the fact that the DTMs often
applied are used to achieve other economic objectives such as pursuit of optimal resource
allocation, equity assurance among others which are at variance with revenue generation
purposes.
Finally, effective tax administration is needed to improve tax compliance and enhance
revenue collection so those who evade tax should be punished severely to serve as
deterrent for others; the punishment should include a fine that will cover cost of recovery
and a penalty or even a recall of mining licenses.
6.1.2 Specific Recommendations
It is desirable to have a built-in flexibility in the tax structure that will enable revenue to
increase by a greater proportion relative to an increase in income thereby facilitating an
increased role of the public sector in the economy. The suggestions below would help
achieve this objective and improve the yield of the tax system in Ghana.
On personal income tax, the study found out that the tax system was inelastic over the
period under consideration. It is recommended that a reduction in allowances and fringe
benefits as well as policies designed to improve coverage and assessment of income tax
University of Ghana http://ugspace.ug.edu.gh
103
would improve the elasticity as they would increase the share of the tax base to GDP. It is
argued that the policy where expatriates whose incomes enjoy higher reduced taxeswas put
in place to make the mining sector more attractive to foreign investment into the sector
(Akbazaa, 2010). The sector is now more investor friendly as its share in FDIs being over
70% can attest to. It is therefore recommended that, a proportion of expatriates‘ incomes
be taxed to also increase the share of the tax base to GDP.
On company taxes, there is the need to reduce the capital consumption allowances which
reduces the base (net profit after tax). Currently, mining companies are allowed to write
off 80% of what they consider as prospecting, exploration and development costs.
Considering that the mining sector is a capital intensive one, these depreciation allowances
afford companies the opportunity to carry forward losses sometimes beyond ten years, thus
providing virtual tax holidays coupled with the fact that surface mines have an average
lifespan of between 10-15 years, most companies may deplete their resources without
paying taxes. A fiscal regime for capital allowance of say 20% for five years of mining
companies should be implemented.
On royalty taxes, the government does not maximise its revenue due to collection
difficulties. Royalty is charged as a percentage of the total value of minerals won,
however, inconsistencies in valuation of minerals won makes revenue tracking more
difficult. Similarly, lack of uniformity in pricing of gold and other mineral commodities
produced by mining companies has led to variation in the computation of royalties, while
the application of different exchange rates regimes by mining companies for the payment
of mineral royalties have produced distortions in the computation of royalty. Fiscal
arrangements that allow for companies to defer or delay royalty payment with permission
University of Ghana http://ugspace.ug.edu.gh
104
from the sector minister should be scrapped. Such requests are commonplace and the
granted delay or deferred payment of royalties has often affected government projected
revenue streams. Since there is apparent lack of capacity for tax authorities to apply sliding
scale tax rates, it is recommended that fixed royalty rates in the region of 5-6% of total
value of mine output be used.
It is important for the tax authorities to identify new sources of taxation which are elastic
and can improve the tax performance in terms of revenue generation. The addition of new
sources of taxation would boost the overall tax buoyancy and elasticity of the tax system.
An example is the windfall taxes. Indeed, this tax is not new as prior to 2006, Ghana had
the Additional Profit Tax Law, 1985 (PNDC Law 122) on her statutes but had apparently
never been applied. An installation of windfall taxes will increase revenues extracted from
this sector because massive profits are being made by mining companies attributable to for
example to gold price increments of over 137% from 2006 to 2010 (US$600 to US$1421)
which was not matched with proportionate increases in operational cost.
A review of the principle of ring-fencing will also boost revenue generation. Ring-fencing
is aimed at preventing companies undertaking a series of projects from deducting costs
from new projects against profitable ventures yielding taxable income. This ensures that
companies‘ ability to evade tax as they start new projects whether feasible or not is
curtailed. With regards to new projects by new entrants into the mining sector, proper
monitoring by tax authorities should curb this act.
University of Ghana http://ugspace.ug.edu.gh
105
6.3 Limitations of the Study
The major limitations of the study include
The unavailability of definite tax bases which, consequently called for the use of
proxies in the estimations.
The study only estimated buoyancy and elasticity for the major taxes and not the
minor taxes with small yield (e.g. fees and licenses, withholding and capital gains
taxes) because the revenue effect for the DTMs of the minor taxes is given as a
global figure in the budget statement.
This research only gives aggregate level information reflecting the impact of the
total tax reform on the individual tax systems with respect to their corresponding
bases. It is not capable of disaggregating these impacts into various sources of
change in each tax under consideration such as the impact of changes in the
statutory tax rate or tax administration efficiency among others. Further research is
therefore required to address the foregoing limitations.
University of Ghana http://ugspace.ug.edu.gh
106
REFERENCES
Abdulai, I. (2005). ―Sectoral Analysis of Foreign Direct Investment in Ghana‖.Bank of
Ghana Working Paper WP/BOG-2005/15.
Addy, S. N. (1998). Ghana: Revival of the Mineral Sector. Resources Policy,24(4), pp.
229–239.
Akabzaa, T. M. (2009).Mining in Ghana: Implications for National Economy
Development and Poverty Reduction.In Bonnie Campbell (Ed.)Mining in Africa,
New York: Pluto Express.
Akabzaa, T. &Abdulai, D. (2001).Impact of Mining Sector Investment in Ghana: A Study
of the Tarkwa Mining Region. Washington: SAPRI.
Andersen, S. P. (1973).―Built-in flexibility of sensitivity of the personal income tax in
Denmark‖. Swedish Journal of Economics, 75, pp. 23-38
Ariyo, A. (1997).Productivity of the Nigerian tax system.AERC Research Paper No. 74.
Artus, K. K. (1974). Tax revenue forecasting: A methodological study with application to
Turkey.Studies in Domestic Finance,9(5), pp. 196-220.
Aryee, B.N.A. (2001). Ghana‘s mining sector: its contribution to the national economy.
Resources Policy, 27, pp. 61-75.
Bahl, R. W. (1971).A Regression approach to tax effort and tax ratio analysis.IMF staff
Papers18, pp. 570-607.
Bank of Ghana, (2003).Annual Publications. Retrieved from
http://bog.gov.gh/index.php?option=com_content&view=article&id=113&Itemid
=172
Bank of Ghana.(2003). Report on the Mining Sector. Research Department, 1, issue 2.
Accra: Bank of Ghana.
Bannock, G., Baxter, R. & Rees, R. (1998).Penguin Dictionary of Economics (6th Ed.),
Middlesex: The Penguin Books Ltd.
Bilquess, F. (2004).Elasticity and buoyancy of the tax system in Pakistan.The Pakistan
Development Review, 43(1), pp. 73-93
University of Ghana http://ugspace.ug.edu.gh
107
Bird, R.M. (1978), Assessing Tax Performance in Developing Countries: A Critical
Review of the Literature on Taxation and Economic Development. London:
Toye, J.F.J., Frank Cass and Company Ltd.
Broadway, R., RobertsS. &Shah, A. (1994). Fiscal federalism dimensions of tax reform in
developing countries. (Working Papers Series 1385).Washington D.C: The World
Bank.
Broadway R. & H. Harris.(1977). A characterization of piecemeal second best
policy.Journal of Public Economics, 8, pp. 169—190.
Brafu-Insaidoo, W.G. &Obeng, C.K. (2008).Effect of import liberalization on tariff
revenue in Ghana.African Economic Research Consortium Research paper, 180.
Cairns, R. D. (1985). Reform of exhaustible resource taxation.Canadian Public Policy, 11,
649-58.
Chaudhry, A. H. M. N. (1962).Thepredictability and the flexibility of tax revenues in
Pakistan. Pakistan Development Review.
Chelliah, R. J. &Sheetal C. K. (1974).A note on techniques of adjusting tax revenue series
for discretionary changes.IMF Staff Papers, 3.
Chhibber, A. &Halilzadeh-Shirazi, J. (1988). Public Finance in Adjustment
Programmes.(PRE Working Paper 128). Washington, D.C: The World Bank.
Chipeta, C. (1998). Tax reforms and tax yield in Malawi. AERC Research Paper, 81.
Choudhry, N. N. (1975). A Study of the West Malaysia income tax system 1961-1970.IMF
Staff Papers, 22(2).
Choudhry, N. N. (1979).Measuring the Elasticity of Tax Revenues: A Divisia Index
Approach. IMF Staff Papers, 26(1).
Deaton, A. (1988). Econometric issues for tax design in developing countries. In D.
Newbery & N. Stem (Eds), The theory of taxation for developing countries.
Washington, D.C: The World Bank.
Demirbas, S. (1999). Cointegrationanalysis – causality testing and Wagner‘s law: the case
of Turkey. Paper Presented at the Annual Meeting of the European Public
Choice Society, pp. 1950-1990,
Dickey, D.A. & W. A. Fuller. (1979). Distribution of the estimators for autoregressive
time series with a unit root.Journal of the American Statistical Association, 74,
pp. 427 - 431.
University of Ghana http://ugspace.ug.edu.gh
108
Diamond, P.A. &Mirrlees, J.A. (1971).Optimal taxation and public production, part II:
"production efficiency", and part II: "tax rules". American Economic Review,
61(1& 3), pp. 8-27; pp. 261-49.
Engle, R.F. & Granger, C.W.J. (1987). Cointegrationand error correction: representation,
estimation and testing.Econometrica, 55, pp. 251-76.
Emerson, C. & Lloyd, P. (1983). Improving mineral taxation policy in Australia.
Economic Record, 59(3), pp. 232-44.
Feldstein, M. (1976).On the theory of tax reform.Journal of Public Economics, 6, pp. 77-
104.
Ghana Chamber of Mines.(2011). “Factoids”, The Mining Sector. Accra, Ghana.
Ghana Chamber of Mines.(2008). Annual Report and Financial Highlights, 2008.Accra:
Chamber of Mines.
Gillis, M. (1990). Tax Reforms and Value-Added Tax: Indonesia. In M. J. Boskin& C. E.
McLure, Jr. (Eds), World Tax Reform. San Francisco: ICS Press.
Gilllani, S.F. (1986).Elasticity and buoyancy of federal taxes in Pakistan. The Pakistan
Development Review, 25(2), pp. 163–174.
Goode, Richard, (1993). Tax advice to developing countries: An historical survey. World
Development, 21, pp. 37–53.
Goode, R. (1987). Disappointed expectations and tax re-form. National Tax Journal,
40(2), pp. 123-30.
Harberger, A.C. (1990). Principles of taxation applied to developing countries: What have
we learned? In M. J.Boskin& C.E McLure, Jr (Eds), World Tax Reform. San
Francisco: ICS Press.
Hinricks, H. (1966).A General Theory of Tax Structure Change during Economic
Development. Cambridge, Massachusetts: Havard Law School.
Kaldor, N. (1964), Essays on Economic Policy, Vol. I, London: Duckworth.
Kargbo B.I.B. &Egwaikhide F.O. (2012).Tax elasticity in Sierra Leone: A time series
approach. International Journal of Economics and Financial Issues, 2(4), pp.
432-447.
Khalilzadeh-Shirazi, J & Shah, A. (1995).Tax Policy in Developing Countries,
Washington, D.C: The World Bank.
University of Ghana http://ugspace.ug.edu.gh
109
Khan, M. Z. (1973).Responsiveness of tax yields to increase in national income. Pakistan
Development Review, 11, 22 – 39.
Kusi, N. K. (1998). Tax reform and revenue productivity in Ghana (AERC Research Paper
74). Nairobi: African Economic Research Consortium
Mackenzie, G. A., Orsmond, D. &Gerson, P. (1997).The Composition of Fiscal
Adjustment and Growth, Washington D. C: International Monetary Fund.
King, M. A. (1982b). Welfare Analysis of Tax Reforms Using Household Data.Journal of
Public Economics, forthcoming.
Mansfield, C.Y. (1972). Elasticity and buoyancy of tax system.International Monetary
Fund Staff Paper, 99, 73-93.
McLure, C.E. Jr. 1989.The Taxation of Income from Business and Capital in Colombia.
Durham, N.C: Duke University Press.
Milton, A., Obwana, M. &Ogwapus, M. (2005).Tax Reforms and Domestic Revenue
Mobilization in Uganda. Washington D.C: Global Development Network.
Minerals Commission (2006).Mineral Commission Statistics on Minerals Production.
Accra: Minerals Commission.
Morrisset, J. &Izquierdo, A. (1993). Effects of tax reform on Argentina’s revenues (WPS
1192). Washington D.C: The World Bank.
Mtatifikolo, F. P. (1990).An economic analysis of Tanzania‗s tax performance experiences
since the 1973 tax act.Eastern Africa Economic Review, 6(1).
Muriithi M. K. & E. D. Moyi. (2003). Tax Reforms and Revenue Mobilization in Kenya
(AERC Research Paper 131), Nairobi: AERC.
Musgrave, R. (1988). Tax reform in developing countries. In D. Newbery and N. Stem
(Eds), The theory of taxation for developing countries. Washington, D.C: The
World Bank.
Musgrave, R.A. (1959). The Theory of Public Finance: New York: McGraw Hill.
Osoro, N. E. (1993).Revenueproductivity implications of tax reform in Tanzania. AERC
Research Paper,Iss.no. 20.
Osoro, N. E. (1995). Tax reforms in Tanzania: motivations, directions and implications.
AERC
Research Paper, Iss.no. 38.
University of Ghana http://ugspace.ug.edu.gh
110
Otto, J.,Batarseh M.L. &Cordes J. (2000). Global Mining Taxation Comparative Study
Oxfam America/ISODEC (2009) Ghana‘s big test: Oil‘s challenge to democratic
development. Report, Feb., 2009
Prest, A. R. (1962). The sensitivity of the yield of personal income in the United
Kingdom.Economic Journal, 72.
PriceWaterhouseCoppers (2007).Metals and Mining.Review of global trends in the mining
industry.
Ramsey, E. P. (1927). A contribution to the theory of taxation.Economic Journal, 37(1),
47-61.
Shalizi, Z. & Squire, L. (1988). Tax Policy in Sub-Saharan Africa (World Bank Policy and
Research Series, No.2.). Washington, D.C: The World Bank
Singer, N. M. (1968).The use of dummy variables in estimating the income elasticity of
state income tax revenues.The Review of Economics and Statistics, 52(4), 427-
433.
Slemrod, J. (1990). Optimal Taxation and Optimal Tax Systems. Journal of Economic
Perspectives, 4(1), 157-78.
Sohato, G. S. (1961). Indian Tax Structure and Economic Development, London:
AsianPublishing House.
Stein, N. (1988a).The Theory of Optimal Commodity and Income Taxation:An
Introduction. In D. Newbery and N. Stein (Eds), The Theory of Taxation for
Developing Countries. Washington, D.C: The World Bank.
Tanzi,V. (1988).The impact of macroeconomic policies on the level of taxation (and on the
fiscal balance) in developing countries (IMF Working Papers, WP/88/95),
Washington, D.C.
Teera, J. M. (2002). Tax Performance: A comparative study (Working Paper 01-02) Centre
for Public Economics: University of Bath.
Terkper, S.E. (1994). Ghana: trends in tax reform. Tax Notes International, 9, pp. 1267-
1275.
Thirsk, W. (1995). Lessons from tax reform: An overview. In Khalilzadeh-Shirazi, J.
&Shah, A. (Eds), Tax policy in developing countries. Washington D.C: The
World Bank.
Tsikata, F. (1997). The vicissitudes of mineral policy in Ghana.Resources Policy, 23(1–2),
pp. 9–14.
University of Ghana http://ugspace.ug.edu.gh
111
Twerefou D. K. &Fumey, A. (2010). Buoyancy and elasticity of tax: evidence from
Ghana. Journal of Monetary and Economic Integration, 10(2).
Wilford, S.D. &Wilford, W.T. (1978a). Estimates of revenue elasticity and buoyancy in
Central America. In Toye, J.F.J (Ed.), Taxation and economic development.
London: Frank Cass & Co. Ltd.
World Bank. (1999). World Bank Development Indicators 1998.Washington D.C: World
Bank.
University of Ghana http://ugspace.ug.edu.gh
112
APPENDIX
APPENDIX A: Total Investment Inflow Into The Mining Sector (2000-2010)
YEAR PRODUCING
COMPANIES
(US$)
EXPLORATION
COMPANIES (US$)
SUPPORT SERVICE
COMPANIES (US$)
TOTAL
(US$)
2000 29.91 179.40 22.47 231.78
2001 108.63 145.21 21.69 275.53
2002 110.50 186.44 18.65 315.59
2003 325.69 198.13 21.80 545.62
2004 407.58 207.36 23.39 638.33
2005 543.12 228.50 25.90 797.52
2006 330.36 232.90 23.48 586.74
2007 410.25 235.41 24.56 670.22
2008 466.75 270.72 27.83 765.30
2009 511.00 222.96 28.30 762.26
2010 508.20 231.00 30.80 770.00
TOTAL 3,751.99 2,338.03 266.87 6,358.89
Source: Minerals Commission of Ghana
University of Ghana http://ugspace.ug.edu.gh
113
APPENDIX B: Socio-Economic Contributions Of The Mining Sector
All values are expressed in thousands of US Dollars
2004 2005 2006 2007 2008 2009 2010
Education 487 604 1,121 1,010 1,406 1,259 2,827
Health 550 333 368 566 416 777 1,055
Electricity 141 66 176 459 334 285 526
Roads 75 692 399 609 2,613 1,376 1,459
Water 265 350 20 221 650 285 679
Housing 265 80 1,290 619 687 113 155
Agro-
Industry
2 1,228 387 779 51
Agriculture 9 11 88 744 1,552 610 809
Sanitation 116 66 2,649 263 405 228 196
Resettlement
Action Plan
548 20 2,649 4,503 568 800 1,190
Alternative
Livelihood
Projects
(other)
242 362 880 2,898 993 799 2,215
Others 371 338 845 831 2,004 2842 6,478
Total 3,070 2,923 11,713 13,109 12,406 9,424 17,590
Source: Minerals Commission of Ghana
University of Ghana http://ugspace.ug.edu.gh
114
APPENDIX C: Mineral Revenues From 2000-2010
Data expressed in million US$
YEAR REVENUE
2000 645.99
2001 642.34
2002 694.97
2003 820.04
2004 835.11
2005 901.22
2006 1,328.18
2007 1,366.27
2008 2,410.72
2009 3,067.87
2010 4,037.05
TOTAL 16,749.76
Source: Minerals Commission of Ghana
University of Ghana http://ugspace.ug.edu.gh
115
APPENDIX D: Breusch-Godfrey Serial Correlation LM Test
Global Buoyancy
Tax Equation Long Run Regression
F-Statistic Probability
TTR 0.3087 0.7386
PYTAX 0.1690 0.8464
COTAX 0.2612 0.7734
ROYTAX 0.9976 0.3906
Tax-to-Base Elasticity
Tax Equation Long Run Regression
F-Statistic Probability
TTR 0.3087 0.7386
PYTAX 0.0881 0.7722
COTAX 0.0881 0.7703
ROYTAX 2.0770 0.1677
Base-to-Income Elasticity
Tax Equation Long Run Regression
F-Statistic Probability
CUPY 0.8253 0.4598
MEXP 0.7697 0.4795
University of Ghana http://ugspace.ug.edu.gh
116
APPENDIX E: DATA USED FOR THE STUDY
YEAR COTAX ROYTAX PYTAX TTR MINING
GDP
MINERAL
EXPORT
CUPY
1990 0.28 0.19 0.47 436.16 7.27 0.40
1991 0.08 0.30 0.38 469.64 14.07 0.52
1992 0.05 0.45 0.50 517.36 23.32 0.90
1993 0.44 0.75 0.26 1.45 564.91 47.36 1.24
1994 0.72 1.28 0.48 2.48 593.68 70.58 1.65
1995 2.04 2.09 0.80 4.93 626.83 108.62 2.43
1996 0.92 3.55 1.68 6.15 653.36 128.26 3.44
1997 0.99 3.46 2.50 6.95 689.73 140.96 4.27
1998 1.45 4.98 3.10 9.53 731.78 193.81 5.51
1999 3.11 4.86 2.78 10.76 753.51 404.52 6.59
2000 1.58 11.87 5.92 19.38 764.48 544.28 7.91
2001 2.48 12.74 7.61 22.83 751.14 546.21 12.16
2002 2.35 15.35 10.15 27.84 784.40 655.89 23.63
2003 6.81 19.44 14.10 40.36 820.85 804.33 31.70
2004 10.03 21.57 13.44 45.04 856.41 823.13 38.61
2005 26.99 23.60 19.41 69.99 883.25 951.98 41.38
2006 40.44 31.63 21.65 93.71 1,019.02 1,289.42 67.26
2007 47.42 40.88 34.59 122.89 1168.69 1924.32 91.07
2008 73.55 59.00 47.14 179.70 1205.75 3307.82
2009 124.60 90.42 103.06 318.08 1407.87 3745.03
2010 241.58 144.70 132.47 518.75 650.55 4655.15
Various sources; data used are in million GHS
University of Ghana http://ugspace.ug.edu.gh
117
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
90 92 94 96 98 00 02 04 06 08 10
LTTR
-.1
.0
.1
.2
.3
.4
.5
90 92 94 96 98 00 02 04 06 08 10
D(LTTR)
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
90 92 94 96 98 00 02 04 06 08 10
LCOTAX
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
90 92 94 96 98 00 02 04 06 08 10
D(LCOTAX)
APPENDIX F: Graphs of Test For Unit Root
Graphs of Variables in Levels and at First Difference.
Total Tax Revenue
Company Tax
University of Ghana http://ugspace.ug.edu.gh
118
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
90 92 94 96 98 00 02 04 06 08 10
LMEXP
.00
.05
.10
.15
.20
.25
.30
.35
90 92 94 96 98 00 02 04 06 08 10
D(LMEXP)
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
90 92 94 96 98 00 02 04 06 08 10
LPYTAX
-.1
.0
.1
.2
.3
.4
90 92 94 96 98 00 02 04 06 08 10
D(LPYTAX)
.00
.05
.10
.15
.20
.25
.30
90 92 94 96 98 00 02 04 06 08 10
D(LCUPY)
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
90 92 94 96 98 00 02 04 06 08 10
LCUPY
Current Personal Income
Mineral Export
Pay-As-You-Earn Tax
University of Ghana http://ugspace.ug.edu.gh
119
University of Ghana http://ugspace.ug.edu.gh
120
-.1
.0
.1
.2
.3
.4
90 92 94 96 98 00 02 04 06 08 10
D(LROYTAX)
5.2
5.6
6.0
6.4
6.8
7.2
7.6
8.0
8.4
90 92 94 96 98 00 02 04 06 08 10
LROYTAX
-.4
-.3
-.2
-.1
.0
.1
90 92 94 96 98 00 02 04 06 08 10
D(LGDP)
8.6
8.7
8.8
8.9
9.0
9.1
9.2
90 92 94 96 98 00 02 04 06 08 10
LGDP
Royalty Tax
Gross Domestic Product
University of Ghana http://ugspace.ug.edu.gh