University of Ghana http://ugspace.ug.edu.gh 
UNIVERSITY OF GHANA 
 
COLLEGE OF HUMANITIES 
 
 
 
 
 
 
TRADE MIS-INVOICING AND ABUSIVE TRANSFER PRICING IN 
GHANA’S COMMODITY SECTOR 
 
 
 
 
 
 
BY 
 
ANGELA AZUMAH ALU 
 
(10228352) 
 
 
 
 
 
 
 
 
 
A THESIS SUBMITTED TO THE SCHOOL OF GRADUATE STUDIES IN 
PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD 
OF DEGREE OF DOCTOR OF PHILOSOPHY IN FINANCE 
DEPARTMENT OF FINANCE 
 
JULY 2022 
 
University of Ghana http://ugspace.ug.edu.gh 
DECLARATION 
I, Angela Azumah Alu, do hereby declare that this thesis is a result of my own work and has not 
been presented to any other academic institution. All references in the work have been duly 
acknowledged. I also declare my full responsibility for any shortcomings in the document. 
 
I acknowledge that the value chain analysis and estimates of trade mis-invoicing for gold and 
cocoa were co-authored with Dr. Ama A. Ahene-Codjoe and Dr. Rahul Mehrotra under the 
Curbing IFFs project. 
 
     
 
 
 
 
 
 
 
 
 
 
 
 
 
 
i 
 
University of Ghana http://ugspace.ug.edu.gh 
CERTIFICATION 
We hereby certify that this thesis was supervised in accordance with procedures laid down by the 
University of Ghana.  
 
 
       
Professor Joshua Yindenaba Abor       Date  
(Principal Supervisor)  
 
 
 
Professor Mohammed Amidu       Date  
(Co-Supervisor)  
 
  
Dr. Ama A. Ahene-Codjoe                                                                  Date     
(Co-Supervisor)  
 
 
 
 
 
 
 
ii 
 
University of Ghana http://ugspace.ug.edu.gh 
DEDICATION 
I dedicate this work to everyone who has supported me throughout my educational pursuits, 
especially my parents.   
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
iii 
 
University of Ghana http://ugspace.ug.edu.gh 
ACKNOWLEDGEMENT 
I extend my sincerest appreciation to the Almighty God for the gift of life, family, strength, and 
all the graces needed for this journey; I am eternally grateful to Him for everything.  My greatest 
appreciation goes to my family, especially my parents, for supporting me throughout the process.  
 
Many people and institutions have provided support in one way or another for this thesis. To 
mention everyone by name would be to run the risk of creating a document longer than the thesis 
itself. I therefore extend my sincerest appreciation to everyone who has in one way or the other 
contributed to this work and to my PhD journey. I am very thankful for your support, 
encouragement, prayers and love and say thank you and God bless you. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
iv 
 
University of Ghana http://ugspace.ug.edu.gh 
ABSTRACT 
This thesis contributes to the limited empirical analysis of the volumes of Illicit Financial Flows 
(IFFs) through trade mis-invoicing recorded in exports from Ghana. The thesis examines the 
volumes of illicit financial flows through trade mis-invoicing in five top export commodities of 
Ghana, namely: gold, cocoa beans, cocoa paste, bauxite, and manganese. It answers the questions 
of what the value chains and the risks for IFFs for those commodities are, how much trade mis-
invoicing there is in these important export commodities and what the determinants for trade mis-
invoicing are in those commodities. 
 
The study uses the Partner Country Trade Gap (PCTG) method to measure trade mis-invoicing 
with macro macro-level statistics from the UN COMTRADE Database as an initial step to 
determine the presence of trade mis-invoicing, followed by the Price Filter Method, which has 
been shown to provide more accurate estimates of IFFs due to the use of transaction transaction-
level data. The thesis employs transaction-level data from the Customs Division of the Ghana 
Revenue Authority in generating the estimates of trade mis-invoicing for the Price Filter Method. 
The study ends with a regression analysis of the determinants of IFFs in the selected commodities.  
 
The value chain analyses of the commodities show that risks for IFFs arise mainly from transfer 
pricing due to multinational firms’ international trade operations. Other risk factors include artisanal, 
small-scale, and informal firms, regulatory infrastructure for verifying export valuation and transit 
trade from neighbouring countries. 
 
v 
 
University of Ghana http://ugspace.ug.edu.gh 
The estimates of trade mis-invoicing generated using the transaction level data from Ghana 
Customs with the price filter methods indicate that: 18.87% of bauxite exports, 11% of gold 
exports, 7.2% of cocoa paste exports, 1% of cocoa bean exports and 0.65% of manganese exports 
were undervalued while 0.2%, 0.9% 3.2%, 4.7% and 2.4% of gold, cocoa beans, cocoa paste, 
bauxite and manganese exports respectively were over-valued from 2011 to 2017.  Overall, the 
results indicate the presence of trade mis-invoicing in Ghana’s commodity exports.  
 
 In analyzing the determinants of trade mis-invoicing in the selected commodities, the study finds 
that for undervaluation, GDP growth, Current Account Balance and the Exchange Rate have some 
influence while the Current Account Balance, Inflation, Tax Differential, GDP growth, Transfer 
Pricing Risk and Transfer Pricing Rank have some influence in the overvaluation observed.   
 
The study has the following policy implications. First, it recommends that policymakers need to 
prioritize the development of institutional expertise to map, control and block the sources of the 
resulting tax base erosion as a result of trade mis-invoicing. Second, there is a need to improve 
data collection capacity of the various institutions engaged in the export of these commodities, 
greater co-operation among the various institutions in these sectors to reconcile data collected and 
constant skills improvement of personnel of these units. Finally, information and communication 
technology tools, especially computers, relevant software and access to critical databases also need 
to be upgraded to match those of the private sector actors in order to ease tax assessments and 
payments tracking. 
 
vi 
 
University of Ghana http://ugspace.ug.edu.gh 
TABLE OF CONTENT 
 
DECLARATION ............................................................................................................................. i 
CERTIFICATION .......................................................................................................................... ii 
DEDICATION ............................................................................................................................... iii 
ACKNOWLEDGEMENT ............................................................................................................. iv 
ABSTRACT .................................................................................................................................... v 
Table of content ............................................................................................................................ vii 
List of Tables ............................................................................................................................... x 
CHAPTER ONE ............................................................................................................................. 1 
INTRODUCTION .......................................................................................................................... 1 
1.1 Background ........................................................................................................................... 1 
1.2 Statement of the Problem ...................................................................................................... 8 
1.3 Research Objectives ...................................................................................................... 12 
1.4 Structure of the Thesis......................................................................................................... 13 
CHAPTER TWO .......................................................................................................................... 14 
LITERATURE REVIEW ............................................................................................................. 14 
2.1 Introduction ......................................................................................................................... 14 
2.2.1 Mercantilism and Standard Theory .............................................................................. 14 
2.2.2 The Classical Trade Theory and Factor Proportion Theory ........................................ 15 
2.2.3 The Heckscher–Ohlin (H-O) Theory ........................................................................... 17 
2.2.4 The Market Imperfections Theory ............................................................................... 19 
2.2.5 The International Production Theory ........................................................................... 20 
2.2.6 The Internalisation Theory ........................................................................................... 20 
2.2.7 Effects of International Trade ...................................................................................... 22 
2.3 Trade Mis-invoicing/Sources and Types of Trade Mis-invoicing ...................................... 27 
2.3.1 The Arm’s Length Principle ........................................................................................ 28 
2.4 Measuring Trade Mis-invoicing .......................................................................................... 33 
2.4.1 Trade Gaps/Trade Discrepancy.................................................................................... 33 
2.4.2 The Partner Country Trade Gap (PCTG) Method ....................................................... 34 
2.4.3 Price Filter Methods (free-market price filter and interquartile range price filter) ...... 37 
2.5 Determinants of Trade Mis-invoicing ................................................................................. 42 
vii 
 
University of Ghana http://ugspace.ug.edu.gh 
2.6 Abusive Transfer Pricing/Transfer Mispricing ................................................................... 48 
2.7 Overview of Anti Money Laundering ................................................................................. 51 
2.8 Chapter Summary ................................................................................................................ 55 
CHAPTER THREE ...................................................................................................................... 57 
METHODOLOGY ....................................................................................................................... 57 
3.1 Introduction ................................................................................................................... 57 
3.2 Value Chain Analyses and Identification of Risks for IFFs in Selected Commodities ...... 57 
3.3 Measuring Illicit Financial Flows ....................................................................................... 61 
3.4 The Partner Country Trade Gap (PCTG) Method ............................................................... 64 
3.5 The Price Filter Method ...................................................................................................... 65 
3.5.1. Arm’s Length Price Range for Gold Exports ............................................................. 68 
3.5.2. Arm’s Length Price Range for Cocoa Beans .............................................................. 71 
3.5.3 Interquartile Range Price Filter Analysis ..................................................................... 74 
3.6 Data Sources ........................................................................................................................ 76 
3.6.1 UN COMTRADE Database ......................................................................................... 77 
3.6.2 Data from Ghana Revenue Authority-Customs Division ............................................ 78 
3.6.3 The Harmonized Commodity Description and Coding System (HS Code) ................ 78 
3.6.4 Free-Market Commodity Price Data: 2011–2017 ........................................................ 80 
3.6.5. Metal Focus Gold Doré Flows Service Database: 2019 ............................................. 80 
3.7 Determinants of Trade Mis-invoicing ................................................................................. 82 
3.7.1 Variables Used for Regression Analysis ...................................................................... 84 
3.8 Chapter Summary ................................................................................................................ 87 
CHAPTER FOUR ......................................................................................................................... 88 
DISCUSSION OF RESULTS ...................................................................................................... 88 
4.1 Introduction ......................................................................................................................... 88 
4.2 Value Chain Analyses and Risks for IFFs .......................................................................... 88 
4.2.1 Gold Value Chain and Risks for IFFs .......................................................................... 89 
4.2.2 Cocoa Sector ................................................................................................................ 95 
4.2.3 Bauxite ....................................................................................................................... 100 
4.2.4 Manganese Value Chain and Risks for IFFs.............................................................. 104 
4.3 Estimates of Trade Mis-invoicing ..................................................................................... 107 
4.3.1 Estimates of Trade Mis-invoicing in Gold (HS 7108.13.1000&7108.12.0000) ........ 108 
4.3.2 Estimates of Trade Mis-invoicing in Cocoa Beans (HS: 1801.00.1100) ................... 110 
viii 
 
University of Ghana http://ugspace.ug.edu.gh 
4.3.3 Estimates of Trade Mis-invoicing in Cocoa Paste (HS: 1803.20.0000). ................... 112 
4.3.4 Estimates of Trade Mis-invoicing in Bauxite (HS:2606.00.0000) ............................ 115 
4.3.5 Estimates of Trade Mis-invoicing in Manganese (HS: 2602.00.0000)...................... 117 
4.4 Determinants of Trade Mis-invoicing ............................................................................... 119 
4.4.1 Descriptive Statistics .................................................................................................. 119 
4.4.2 Correlation Analysis .................................................................................................. 120 
4.4.3 Regression Analysis ................................................................................................... 121 
4.4.2 Robustness Checks: Quantile Regression .................................................................. 124 
4.5 Chapter Summary .............................................................................................................. 129 
CHAPTER FIVE ........................................................................................................................ 130 
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS.............................................. 130 
5.1 Introduction ....................................................................................................................... 130 
5.2 Summary ........................................................................................................................... 130 
5.3 Conclusion ......................................................................................................................... 136 
5.4 Contributions to Knowledge ............................................................................................. 137 
5.5 Recommendations ............................................................................................................. 138 
5.6 Limitations and Directions for Further Research .............................................................. 140 
REFERENCES ........................................................................................................................... 141 
APPENDICES ............................................................................................................................ 157 
Appendix 1: Results of Hausman Tests .................................................................................. 157 
Appendix 2: Descriptive Statistics .......................................................................................... 158 
Appendix 3: Endogeneity test ................................................................................................. 158 
 
 
 
 
 
 
 
 
 
ix 
 
University of Ghana http://ugspace.ug.edu.gh 
LIST OF TABLES 
Table 3.1: List of Institutions Contacted....................................................................................... 60 
Table 3.2 Main Gold Mines in Ghana, Ownership and Gold-Silver Content in Production ........ 81 
Table 3.3: Variables used for regression analysis ......................................................................... 85 
Table 4:1 Undervalued and Overvalued Exports-Gold .............................................................. 108 
Table 4.2: Undervalued and Overvalued Exports – Cocoa Beans .............................................. 110 
Table 4.3: Undervalued and Overvalued Exports – Cocoa Paste ............................................... 114 
Table 4:4 Undervalued and Overvalued Exports – Bauxite ....................................................... 115 
Table 4.5: Undervalued and Overvalued Exports – Manganese ................................................. 117 
Table 4.6 Descriptive Statistics ................................................................................................... 119 
Table 4.7 Correlation Table ........................................................................................................ 120 
Table 4.8 Ordinary Least Squares (OLS) Regression ................................................................. 121 
Table 4.9 Random Effects Regression ........................................................................................ 122 
Table 4.10 Quantile Regression for Overvaluation using Transfer Pricing Risk ....................... 126 
Table 4.11 Quantile Regression for Overvaluation using Transfer Pricing Rank ...................... 127 
Table 4.12 Quantile Regression for Undervaluation using Transfer Pricing Risk ..................... 128 
Table 4.13 Quantile Regression for Undervaluation using Transfer Pricing Rank .................... 129 
 
 
 
x 
 
University of Ghana http://ugspace.ug.edu.gh 
CHAPTER ONE 
INTRODUCTION 
1.1 BACKGROUND 
It is important to build robust domestic fiscal and financial systems and improve domestic revenue 
mobilisation to finance the development agenda of African countries sustainably (African 
Economic Outlook, 2010). Domestic resource utilization can be strengthened by expanding the tax 
base and enhancing tax management in developing countries. To improve tax collection, leakages 
within the tax collection system should be plugged, especially precarious leakages that have been 
exposed recently; that is, the loss of tax revenue due to Illicit Financial Flows (IFFs). For example, 
the High-Level Panel on IFFs estimated that Africa loses approximately USD 50 billion annually 
through trade mis-invoicing (UNECA, 2015). 
 
Similarly, Global Financial Integrity in 2017 estimated that trade mis-invoicing from developing 
countries could potentially equal USD 1 trillion. The funds lost through IFFs are a potential source 
of revenue mobilisation from Africa’s domestic resources, which could potentially have positive 
effects on Africa’s development agenda (Abor 2014). This is especially true of recent international 
economic trends, which signal that dependence on development aid is no longer a sustainable 
choice for developing countries (United Nations Economic Commission for Africa, 2015).  
 
In general, IFFs refer to illicit or illegal transfers of funds or capital from one country to another. 
The Organisation for Economic Co-operation and Development (OECD, 2014) defines IFFs as 
financial resources generated outside of a country in violation of national or foreign legislation by 
processes and procedures or through crime. IFFs can include funds of criminal origin such as 
1 
 
University of Ghana http://ugspace.ug.edu.gh 
proceeds from crime (including corruption), funds with criminal intent such as bribery, terrorist 
funding and conflict funding, tax evasion funds, payments to, from, or for organisations subject to 
financing penalties and money-laundering / counter-terrorist financing payments. The World Bank 
(2016) defines IFFs as “money illegally earned, transferred, or used that crosses borders”. The 
term encapsulates international funds transfers generated through bribery, tax evasion and illegal 
markets. 
 
Similarly, according to Kar and Freitas (2012), IFFs refer to money received, moved or used in 
breach of current legislation. Some studies approach the problem by describing IFFs not by 
behaviour but by adverse effects. For example, Blankenburg and Kahn (2012) consider any 
financial activity which damages the economy as part of IFFs, creating a slightly vague meaning 
which could include many borderline activities between legal and illegal. 
 
The term IFFs is a relatively new concept (World Bank, 2016) which emerged to bring together 
previously unconnected issues concerning resources, strategies, illegal activities aimed at moving 
funds and properties across national boundaries in contravention of foreign or national laws 
(Goredema, 2011). There is still no consensus about what actions these and other illegal activities 
represent (World Bank, 2016).  Despite the general agreement on using this relatively new term, 
there is a lack of consensus in literature regarding the sources of the IFFs. Most researchers agree 
that money earned from crime (illegal activities) or money legally earned but illegally used, such 
as terrorist funding or illegal transfers, should be included within this term. The current literature 
includes numerous sources of money originating from illegal activities such as corruption, illicit 
activities (including organised crimes), and unfair trading practices such as customs bribery, 
2 
 
University of Ghana http://ugspace.ug.edu.gh 
mispricing and tax avoidance (Baker, 2014). There are also other classifications of IFFs. Jansky 
(2013), for instance, splits IFFs into three groups: illegal (criminal) flows individual illicit flows 
and corporate (commercial) illicit flows.  
 
IFFs are of policy importance because it drains developing countries’ capital and tax revenue 
(Reuter, 2017). Over the last 50 years, Africa is estimated to have lost more than USD 1 trillion 
through IFFs (see Kar & Cartwright-Smith, 2010; Kar & Leblanc, 2013). This amount roughly 
equals all official African development funding in the same timeframe. Also, the decisions by 
multinational corporations to transfer profits from high tax jurisdictions to low tax jurisdictions 
leads to reductions in tax revenue for the affected states which are predominantly developing 
economies characterised by high taxes (Shaxson, 2012). Furthermore, by illegally transferring 
these assets away from the government, tax evaders, corrupt officials, or criminals can better 
defend themselves against prosecution. 
 
IFFs, however, do more than just plunder a nation. According to Shaxson (2012), IFFs constitute 
a major threat to the progressive realization of human rights. This is because they threaten every 
fundamental aspect of a country’s growth and development. It is assumed that if the elite and other 
leaders of a nation can keep a lot of their wealth outside the country, they may tend to have less 
incentive to ensure the development of property rights, which is a crucial element of the rule of 
law. Furthermore, when it becomes public knowledge that the wealthy are moving their wealth out 
of the country, broader public trust is eroded.  
 
3 
 
University of Ghana http://ugspace.ug.edu.gh 
According to United Nations Economic Commission for Africa (2015), IFFs are becoming a 
danger to sustainable development and are presently one of the most significant obstacles to global 
growth since they undermine economic growth and trade, also deprive governments of financial 
resources which could have been used in the provision of public goods including healthcare, 
education and infrastructure. In institutionally weak and emerging countries, IFFs are especially 
prevalent and destructive since they appear to exploit and aggravate vulnerabilities in public 
institutions, weaken governance and empower those outside that law.  
 
Trade mis-invoicing is a type of fraudulent activity involving exporters and importers who 
deliberately misreport, in a commercial transaction, the value, quantity or nature of goods or 
services meant to be transferred to or from another destination. Estimates of trade mis-invoicing 
are closely linked to the term IFFs as both were popularised by Global Financial Integrity (GFI) 
(Forstater 2018). Trade mis-invoicing is a major channel for illegal financial flows such as capital 
flight and money laundering, tax avoidance and the elimination of duty. It occurs when the value 
of the transaction declared by the customs agency differs from the value of a non-binding export 
or import transaction (Hong & Pak 2017). There are four basic categories of trade mis-invoicing: 
import under-invoicing, import over-invoicing, export under-invoicing and export over-invoicing. 
Trade mis-invoicing, which is illegal, happens when traders under-invoice or over-invoice export 
or import for the purpose of tax evasion or capital flight. When this happens, the economy is 
stripped of national resources that could otherwise have been used domestically or exported 
overseas for revenue inflows. This causes a decline in economic growth because capital is not 
made available. Additionally, the lack of revenue also implies that the economy does not have the 
available resources to improve social services (Qureshi & Mahmood, 2016). 
4 
 
University of Ghana http://ugspace.ug.edu.gh 
According to Kar and Cartwright Smith (2010), commercial activities of companies (through mis-
invoicing and abusive transfer pricing) account for 65% of IFFs, while criminal activities account 
for 30% and bribery and corruption account for 5% of IFFs. Given these percentages, there is quite 
a focus on trade mis-invoicing and abusive transfer pricing. According to the High-Level Panel 
report on Africa’s IFFs, more than 54% of Africa’s IFFs are due to trade mis-invoicing, estimated 
at USD 407 billion over the ten-year period from 2001 and 2010 (UNECA, 2015). Global Financial 
Integrity (GFI) also estimates that trade mis-invoicing makes up over 80% of illicit outflows 
(Baker 2014). Trade mis-invoicing is the primary method used in IFFs, including capital flight 
(Kar & Freitas, 2012). It is therefore important to correctly estimate the degree of trade mis-
invoicing to determine its contribution to IFFs and losses in tax and duty income in a country of 
interest. 
 
In recent years the awareness of trade mis-invoicing has grown, thereby necessitating research to 
provide further insights into the scale of IFFs and their implications for lost revenue within the 
framework of the current global agenda on development funding and mobilisation of domestic 
resources. Most empirical studies (see de Boyrie, Nelson & Pak, 2007; Baker 2014) have focused 
on the analysis of estimates of illegal capital flows from developing countries through import over-
invoicing and export under-invoicing, vis-à-vis their developing country trading partners and 
worldwide often generating aggregate figures. However, the measurement of trade mis-invoicing 
remains a challenge.   
 
Trade mis-invoicing has mainly been measured using the Partner Country Trade Gap (PCTG) 
Method, developed by Bhagwati (1964a). This methodology operates on the principle of a trade 
5 
 
University of Ghana http://ugspace.ug.edu.gh 
discrepancy. In international trade transactions, the data reported by one country is expected to be 
the same as the data reported by the other partner country after adjusting for transportation and 
insurance expenses. In certain instances, this does not occur for specific reasons. Choi and 
McGuaran (2018) noted that trade differences in trading data could occur simply on statistical and 
logistical grounds. For example, the difference in CIF/FOB ratio, incoterms, attribution of trade 
partner, Foreign Exchange rate, time, low-value thresholds between trade partners and any 
adjustments to the import customs value to be made under Article 8 of the Agreement on 
Implementation of Article VII of the GATT 1994. Also, discrepancies could arise in international 
trade transactions from transit trade, commodity processing and re-exports (Nitsch, 2016). 
Although trade mis-invoicing has also been identified as one reason for discrepancies in bilateral 
trade data, the estimates generated from the PCTG method have mainly been criticized for being 
“guesstimates” and for over-estimating the magnitudes of trade mis-invoicing (Nitsch 2016; Collin 
2020; Marur 2019).  
 
In addition, trade mis-invoicing of funds may be influenced by factors such as the real interest rate 
that reflects the expected return on capital spending in the economy as well as the corporate tax 
rate (Tandon & Rao, 2017). Also, the economic conditions of either the exporting country or the 
importing country may cause trade mis-invoicing. For instance, if the exporting country’s interest 
rate is higher than the rest of the world, trade mis-invoicing may help to encourage inflows into 
the nation. 
 
A similar but related thread is abusive transfer pricing. Thomas (1971) pioneered the discussions 
on transfer pricing. The transfer price is the transaction rate between two entities belonging to the 
6 
 
University of Ghana http://ugspace.ug.edu.gh 
same business group. For example, on behalf of its Ghanaian subsidiary, a South African company 
might supply mining equipment or machinery and charge a service fee. The price agreed upon is 
the “transfer price,” and the process for setting it is referred to as “transfer pricing” (Readhead, 
2016). Transfer pricing affects transaction pricing between associated companies. An associated 
business in this sense refers to a parent, a trustee or a partnership where the entity is a partnership 
that controls at least 50% of the rights to the income or capital of the partnership and an enterprise 
managed by the person alone or with other associates. However, this transfer price can sometimes 
be abused, and this is referred to as abusive transfer pricing.   
 
According to African Economic Outlook (2012), a provision for many Multinational Companies 
(MNC) subsidiaries and countries in developing and emerging markets, which requires strict 
secrecy, is one of the most significant challenges to the natural resource sector. This creates an 
atmosphere of secrecy where accounting laws can be abused with tax avoidance. MNCs in the 
field of natural resources are also more discreet than others. Accordingly, there is a common 
assumption that natural resource-rich developing countries do not benefit from the amount of tax 
they can earn from the industry itself. This goes deeper than a clear explanation of the lack of 
potential of developing countries. 
 
In developing countries, there is an absence of effective and complementary regulations on 
transactions, thus enabling the climate for coercive pricing by manipulating the transfer price. 
Although the problems of price transfer manipulation affect all industries, mining natural resources 
can be particularly troublesome and can also be a refuge for corruption and other illegal activities 
(McLennan, 2012). 
7 
 
University of Ghana http://ugspace.ug.edu.gh 
Another similar concept is profit-shifting. Generally, the literature defines profit shifting as a 
systemic benefit shift among MNC members to decrease global taxes payments (Schimanski, 
2018). In the light of UN sustainable development goals 16.4 and 17.1, a reduction in profit-
shifting is also considered a highly important area of study, considering that a significant reduction 
in illegal financial flows could enhance the mobilisation of domestic income (Schimanski, 2018). 
 
Given the importance of tax revenue to developing countries such as Ghana in meeting their 
developmental goals, it is very necessary to have accurate estimates of the magnitudes of IFFs 
through trade-based mis-invoicing and the resultant revenue and economic losses in order to 
understand what actions need to be taken to curtail the loss of revenue through IFFs. In addition, 
there is a need to understand the factors influencing trade-based mis-invoicing.  
 
1.2 STATEMENT OF THE PROBLEM 
IFFs are a bane of developing countries because they divert critical resources for developmental 
projects. This has become even more critical as traditional sources of development finance, such 
as Official Development Assistance (ODA), are dwindling following the global financial crisis 
and other challenges facing developed countries. For instance, aid from OECD countries decreased 
by 2.7% in 2011 compared to 2010 because of the global financial crisis. Bilateral aid to sub-
Saharan Africa also fell by 0.9% (28 billion USD in real terms). In comparison, bilateral ODA to 
the group of Least Developed Countries fell by 8.9% (USD 27.7 in real terms) billion during the 
same period (Balma, Bawumia & Ncube, 2018).  
 
Reducing IFFs could make more resources available for financing the Sustainable Development 
Goals (SDGs). This is particularly important because most SDG funding is expected to come from 
8 
 
University of Ghana http://ugspace.ug.edu.gh 
domestic resources.  Indeed, one of the SDGs (goal 16:4) is to reduce illegal financial flows 
significantly, increase stolen asset recovery and returns and counter all types of organised crime 
by 2030. It also seeks to improve domestic revenue mobilisation (Goal 17.1).  A study by the 
United Nations Economic Commission for Africa (2015) found that IFFs tend to be more prevalent 
in the extractives and commodity sectors. This makes it imperative to accurately identify the 
sources of IFFs and measure their magnitude to be able to introduce the needed policy measures 
to curb IFFs and improve domestic revenue mobilisation. 
 
Although seemingly neglected in IFFs studies on Africa, Ghana is a country of interest as far as 
IFFs are concerned because of its rich natural resources and fast-growing economy. The country 
is endowed with natural resources such as cocoa, gold, bauxite, manganese, petroleum, cashew, 
timber and coffee. Ghana is also the second leading producer of cocoa in the world and the leading 
producer of gold in Africa. The country relies heavily on export revenue from these commodities 
and others, and as such is vulnerable to trade mis-invoicing incidents. In 2017, the mining sector 
contributed 16.30% to government revenue in the form of domestic tax and 43% of total 
merchandise exports (Ghana Chamber of Mines, 2018). According to Abor (2014), Ghana is 
reported to be among the top ten countries in Africa with high levels of IFFs. However, only a few 
studies have focused on Ghana, including Baker (2014), which estimated trade mis-invoicing in 
Ghana using the PCTG method with data from UN COMTRADE. Hence this thesis focuses on 
five of Ghana’s top export commodities. Gold, bauxite and manganese are among Ghana’s top 
export commodities and as extractives, tend to be vulnerable to IFFs (UNECA 2015). Cocoa beans 
and cocoa paste are also chosen because of their economic significance: Ghana is the second 
9 
 
University of Ghana http://ugspace.ug.edu.gh 
leading producer of cocoa beans in the world while cocoa paste is Ghana’s top Non-Traditional 
Export commodity. 
 
Some studies have measured trade mis-invoicing using varying approaches with limitations. They 
focused on measuring trade mis-invoicing using the Partner Country Trade Gap (PCTG) Method. 
The PCTG Method, first developed by Bhagwati (1964b), is based on the concept that what one 
country exports should equal what is imported by the partner country after adjusting for 
transportation and insurance costs. Any discrepancy is assumed to be indicative of trade mis-
invoicing.  
 
In 2015, United Nations Economic Commission for Africa estimated IFFs losses in five African 
countries, while a study by Global Financial Integrity also estimated trade mis-invoicing in five 
developing countries using the PCTG method (Baker 2014). Other studies include Marur (2019) 
generated PCTG estimates for cocoa and gold from Ghana and copper and coffee from Laos; 
Mehrotra and Carbonnier (2020) for Switzerland as a trading hub and Nolintha et al. (2020) for 
Laos as an exporting country. Baker (2014) estimated that Ghana loses about USD 1 billion 
annually through trade mis-invoicing using the Partner Country Trade Gap (PCTG) Method.  
 
Typically, studies that have used the PCTG method of 10% for transportation costs which is one 
of the acceptable reasons for trade gaps. However, the 10% has been criticised as being arbitrarily 
chosen (see De Wulf, (1981); Nitsch, (2016); Hong &Pak, (2017); Marur (2019)), hence the need 
for better, more informed estimates. According to Reuter (2012), GFI and other institutions' work 
is intended to encourage academic researchers, the International Monetary Fund (IMF), the World 
10 
 
University of Ghana http://ugspace.ug.edu.gh 
Bank, and other interested parties to collect more reliable and complete data in order to develop 
more robust evaluation methods. National trade data estimates are considered inadequate because 
they do not fully present the volumes attributable to trade mis-invoicing based on the limitations 
of the PCTG methodology. As such, commodity-level data, or ideally transaction-level 
information, are needed for generating the right volumes of trade mis-invoicing (Cobham &Jansky 
2017).  
 
The criticisms of the PCTG method led other authors such as Pak and Zdanowicz (1994) and Hong 
et al. (2014) to develop the price filter method to measure trade mis-invoicing at the micro-level 
using transaction-level data. The methodology incorporates the arm’s length price range which 
utilizes free-market prices of the commodities as the arm’s length price and the Inter-Quartile 
Range (IQR) filter, which utilises a statistical range as the arm’s length price. 
 
The free-market arm’s length price methodology is based on the World Trade Organization’s 
Transaction Value methods for customs valuation and the Comparable Uncontrolled Price (CUP) 
method for transfer pricing analysis of trade between related firms (Platform for Collaboration on 
Tax, 2017; United Nations, 2017). This methodology has been used in some studies to measure 
trade mis-invoicing. These studies include Pak and Zdanowicz (1994), Hong et al. (2014), and 
Hong and Pak (2017). However, these studies used a 10% filter to generate their estimates of trade 
mis-invoicing using the arm’s length price range of the price filter method. Hong et al. (2014), in 
a study on the US Banana trade, recommended the choice of 10% as an area for future research. 
In this thesis, extensive commodity research was used to choose the price filters for the arm’s 
length price range rather than the arbitrary 10% commonly used.  
11 
 
University of Ghana http://ugspace.ug.edu.gh 
The IQR filter presupposes those transactions falling between the 25th and 75th percentiles of a 
commodity's distribution represent normal prices. In contrast, any transactions outside of this range 
would be considered trade mis-invoicing, i.e. undervaluation or overvaluation. This method is 
suitable for commodities traded without reference to a single benchmark price. Since the IQR is 
calculated endogenously using the observed distribution of export prices, this method directly 
assumes the presence of under and over-valuation in the trade statistics. However, it provides 
useful insights from a risk analysis perspective by identifying exporters and trade partners who 
consistently appear in the extreme tails of the products’ observed price distribution (Zdanowicz et 
al., 1999; Pak et al., 2003; De Boyrie et al., 2005; Hong & Pak, 2017). The OECD Transfer Pricing 
Guidelines recommend using the interquartile range (OECD, 2017). In addition, the thesis 
examines the determinants of trade mis-invoicing in Ghana.  
 
To summarize, this study adds to literature by examining trade mis-invoicing in Ghana from a 
commodity-level perspective as follows; first, by analyzing the value chains of the commodities 
to identify the risks for IFFs in the commodities, second using different methods to measure trade 
mis-invoicing with data from the UN COMTRADE and the Ghana Revenue Authority, and finally 
examining the determinants of trade mis-invoicing. The commodities chosen are gold, cocoa 
beans, cocoa paste, bauxite and manganese. 
 
1.3 RESEARCH OBJECTIVES 
The main objective of this study is to examine trade mis-invoicing and transfer pricing in 
Ghana’s commodity markets.  
Specifically, the objectives of the study are: 
12 
 
University of Ghana http://ugspace.ug.edu.gh 
i. to analyze the value chains of the gold, cocoa, bauxite and manganese commodity sectors 
and identify the risks for trade mis-invoicing and transfer pricing along the value chains 
ii. to estimate trade mis-invoicing in Ghana’s commodity sector using the Partner Country 
Trade Gap Method and the Price Filter Method, and 
iii. to investigate the determinants of trade mis-invoicing in the commodity sector in Ghana. 
 
1.4 STRUCTURE OF THE THESIS 
The thesis is structured into five chapters. Chapter one presents the background to the study, the 
problem statement and research objectives of the study. Chapter two is the literature review 
chapter. It reviews relevant extant literature while chapter three presents the empirical 
methodologies and the data sources used. Chapter four presents the results and analysis obtained 
from the study and chapter five provides the summary of findings, conclusions, contributions of the 
study, policy recommendations, as well as limitations and suggestions for further research. 
 
 
 
 
 
 
 
13 
 
University of Ghana http://ugspace.ug.edu.gh 
CHAPTER TWO 
LITERATURE REVIEW 
 
2.1 INTRODUCTION 
This chapter reviews literature relevant to the study. The literature discussed in this chapter 
includes a theoretical overview of international trade. It further examines various concepts such as 
trade discrepancy and the arm’s length principle, explains trade mis-invoicing and the various 
ways of measuring trade mis-invoicing and ends with the determinants of trade mis-invoicing. 
Various sub-themes are discussed to appreciate the state of the extant literature on IFFs while 
making way for the useful additions made by this study. 
 
2.2.1 Mercantilism and Standard Theory 
International trade before the nineteenth century strikingly displays a movement of the factors of 
production underlying, requisite to and proceeding out of the anticipation of profits to be made by 
international extension of markets and raw materials. Mercantilism was the main economic 
structure in many industrial nations from the sixteenth to the eighteenth centuries. The Mercantilist 
approach to foreign trade believed that a country's prosperity was mainly measured by its 
ownership of metals such as gold and silver. These metals were obtained by promoting exports 
and encouraging metal discoveries in the Americas, thus suppressing imports by introducing 
unnecessary tariffs (Peukert, 2012). After nearly 300 years of insecurity and economic failure, 
what became known as the Standard Theory of International Trade severely criticised mercantilism 
(Wilson, 1959). The Standard Theory's development can be traced back to 1776 and 1817, when 
Adam Smith's Wealth of Nations (Smith, 1776) and David Ricardo’s Principles of Political 
Economy and Taxation (Ricardo, 1817a) were published. These two publications signalled the 
14 
 
University of Ghana http://ugspace.ug.edu.gh 
formulation of a theory of free trade, given England's exceptional success in the respective fields 
of industry and trade (Sen, 2010). 
 
The Standard Trade Theory relates merchandise to actual exchange rate fluctuations using a 
straightforward, common-sense method in which all other variables are kept constant. The value 
and amount of trade are also affected by exchange rate fluctuations. As a result, if the home 
country's real exchange rate rises, resulting in real depreciation, households receive less imported 
goods in exchange for a unit of domestic goods and services. This implies that a unit of imported 
goods would yield more domestic goods. Domestic households will ultimately buy few imports, 
while foreign households will buy comparatively more domestic goods. Finally, the greater the 
actual exchange rate of the home country, the greater the country's trade surplus (Zhang, 2008).  
 
Lerner (1944) expanded traditional trade theory by incorporating the demand price elasticities of 
imports and exports as instrumental elements in calculating the impact of actual exchange rate 
fluctuations on the trade balance. As a result, an increase in exports and a decline in imports as a 
result of a depreciation in the real exchange rate does not always suggest a reduction in the trade 
balance deficit. According to Lerner (1944), the trade balance is concerned with the actual prices 
of tangible products rather than their number. 
 
2.2.2 The Classical Trade Theory and Factor Proportion Theory 
According to classical trade theory, the degree to which a country exports and imports is related 
to its trading pattern with other countries. That is, countries will benefit if each devotes resources 
to the development of products and services in which they have a comparative advantage (Ricardo, 
15 
 
University of Ghana http://ugspace.ug.edu.gh 
1817b; Smith, 1776). As a consequence, classical trade theory essentially explains the situation in 
which a country produces goods and services in which it has an advantage for domestic 
consumption and then exports the surplus. As a result, it makes sense for countries to export 
products and services in which they have a comparative advantage. Comparative 
advantages/disadvantages can result from differences in factors such as resource endowment, 
labor, capital, technology, or entrepreneurship between countries. Thus, according to classical 
trade theory, the basis for foreign trade can be traced back to disparities in commodity 
characteristics and resource endowments, which are centered on domestic differences in natural 
and acquired economic advantages.  
 
In comparison to classical trade theory, factor proportion theory may clarify the disparities in 
benefit shown by trading countries. As said by this theory, countries will strive to produce and 
export goods and services that harness large amounts of abundant production factors that they 
possess, while importing goods and services that involve large amounts of relatively scarce 
production factors (Hecksher & Ohlin, 1933). 
 
As a result, this principle broadens the idea of economic benefit by taking into account the 
endowment and costs of production factors. However, both of these hypotheses have been shown 
to be ineffective in understanding more recent trends of foreign trade. For example, the 1960s saw 
major technological change and the rise of multinational companies, triggering a demand for new 
theories of foreign trade to represent evolving commercial realities (Leontief, 1966). The product 
life cycle theory of international trade was found to be a valuable framework for understanding 
and forecasting international trade trends as well as multinational business expansion at the time. 
16 
 
University of Ghana http://ugspace.ug.edu.gh 
According to this theory, a trade cycle occurs when a commodity is made by a parent company, 
then by its international subsidiaries, and eventually anywhere in the world where prices are at 
their lowest (Vernon, 1966, 1971; Wells, 1968, 1969). It also explains how a product can begin as 
a country's export and progress through the life cycle to become an import. The essence of the 
international product life cycle is that technological progress and market growth are important 
factors in explaining international trade trends. That is, technology plays an important role in the 
creation and development of new goods, while market size and structure have an effect on the 
degree and form of foreign trade.  
 
2.2.3 The Heckscher–Ohlin (H-O) Theory 
The Heckscher–Ohlin (H-O) theory explains the source of international productivity differences: 
the driving factor that determines comparative advantage and the pattern of international trade. 
Two Swedish economists, Heckscher (1919) and Ohlin (1933) extended the Ricardian trade model 
into what is now known as the Heckscher–Ohlin (H-O) theory by introducing one more input, 
namely, capital, in addition to labour in the Smithian and Ricardian models. Heckscher and Ohlin 
argued that comparative advantage arises from differences in national resources or factor 
endowments. As a result, the more abundant a factor is, the lower its cost, giving the country a 
tendency to follow a production process that heavily relies on the relatively abundant factor. The 
H-O model predicts that countries will export goods that make intensive use of locally abundant 
factors and import goods that make intensive use of locally scarce factors, based on the assumption 
that different commodities need different intensities of factor inputs in their production. In 
summary, capital-rich countries such as the United States of America and other industrial 
17 
 
University of Ghana http://ugspace.ug.edu.gh 
economies should export capital-intensive goods, whereas labour-rich countries such as Ghana and 
other developing economies should export labour-intensive goods (Hill, 2009; Salvatore, 1998). 
 
In effect, the now famous Heckscher-Ohlin paradigm and the analysis of Mundell (1957a) provides 
a reference for studying the interaction of trade integration and capital mobility. In the classical 
Heckscher-Ohlin-Mundell (HOM), two-goods, two-factor framework and free trade lead to factor 
price equalisation with the rest of the world. Once factor price equalisation has materialised, the 
mobility of international capital becomes irrelevant. For example, consider two countries with 
different levels of capital endowment. If the two countries can trade with each other, there is no 
need for capital to flow from the capital-rich to the capital-scare country because rate of return 
disparities can be reduced by trade alone. As a consequence, global trade and capital flows function 
as substitutes. Several contributions, however, have introduced more practical features to the basic 
HOM framework, ranging from technical differences (Kemp, 1966; Jones, 1967) to production 
uncertainty (Helpman & Razin, 1978). Under these changes, trade and factor flows may be 
supplemented, with causality running from foreign capital to trade flows. Markusen (1983) shows 
that the finding of Mundell (1957b) is an exception rather than the rule, since Mundell concludes 
that trade in products and factor flows behave as complements in general by comparing multiple 
models and modeling assumptions. 
 
Furthermore, recent contributions aim to integrate financial frictions into macroeconomic 
dynamics in order to model the trade-finance nexus. For instance, Antràs and Caballero (2009) 
forcefully argue that capital and trade flows are complements. In their model, greater trade 
integration leads to greater capital inflows to the capital-scarce region, as compared to previous 
18 
 
University of Ghana http://ugspace.ug.edu.gh 
research that found a causal correlation between trade and foreign capital flows. Other modeling 
approaches provide a correlation between trade and capital flows and the maturity level of financial 
institutions. Furusawa and Yanagawa (2013), in their study, emphasize the fact that the less mature 
financial institutions are, the more trade and capital flows function as complements. Rose and 
Spiegel (2002) also present a theoretical model based on the argument that countries service their 
external debts out of fear of a drop in trade. As a result, countries prefer to trade more with 
countries with which they have stronger financial links. Other research suggests that transaction 
costs (Portes et al., 2001) or knowledge asymmetries (Hahm & Shin, 2009) contribute to 
complementarity trade in goods and financial assets.  
 
2.2.4 The Market Imperfections Theory 
According to the market imperfections theory, companies are always looking for opportunities. As 
a result, their decision to invest abroad can be viewed as a means of capitalising on capabilities 
that their competitors in other countries lack (Hymer, 1970). Market imperfections for goods and 
factors of production justify firms' capabilities or advantages. According to the principle of perfect 
competition, companies must produce homogeneous goods and have equal access to production 
factors. However, imperfect competition is a fact that is embodied in industrial organisation theory. 
Porter (1985) determines the degree to which companies achieve various types of comparative 
advantages. Nonetheless, market imperfections theory does not justify why the international output 
is regarded as the most attractive method of capitalizing on a firm's advantage. Dunning (1980) 
and Fayerweather (1982) have however addressed this issue and developed what can be described 
as international production theory. 
 
19 
 
University of Ghana http://ugspace.ug.edu.gh 
2.2.5 The International Production Theory 
According to international production theory, a firm's proclivity to begin foreign production is 
determined by the basic attractions of its home country in comparison to the resource implications 
and advantages of locating in another country. This theory makes it clear that not only do resource 
differentials and company advantages play a role in deciding overseas investment practices, but 
that foreign government decisions can have a huge impact on the piecemeal attractiveness and 
entry conditions for firms. A related element of this foreign investment theory is the idea of 
internalisation, which has been thoroughly researched by Buckley (1982, 1988) and Buckley and 
Casson (1976, 1985). 
 
2.2.6 The Internalisation Theory 
Internalisation theory is based on the idea that firms strive to create their own internal markets 
whenever transactions within the business can be completed at a lower cost. Consequently, 
internalisation entails a form of vertical integration in which new operations and activities, 
previously carried out by intermediary markets, are brought under the firm's ownership and 
governance. However, much of the theory used the global corporation as the unit of review and 
ignored the mechanism that led up to that stage of international growth. In response, a more 
complex, process-based viewpoint was needed, as well as recognition of the firm's 
internationalisation. 
 
As Hayes and Abernathy (1980) illustrated in a seminal article, a country's trade deficit 
performance cannot always be described by macroeconomic phenomena. It must be acknowledged 
that the entrepreneur plays a role in explaining a country's foreign trading activities. Given that an 
20 
 
University of Ghana http://ugspace.ug.edu.gh 
economy can consist of many industries housing a diverse range of businesses, it is fair that the 
position of decision-makers within such organizations may, collectively, provide a significant 
contribution to economic performance. Compared to the above-mentioned international trade and 
FDI theories, internationalisation theories attempt to understand how and why a company 
participates in overseas operations and, more specifically, how the complex essence of such 
behavior can be conceptualised. Despite the fact that the word "internationalisation" has been 
widely used, few genuine attempts have been made to provide an operational sense of its meaning. 
Piercy (1981) and Turnbull (1985) describe internationalisation as the outward movement of a 
firm’s operations. This term, however, may be expanded to define internationalisation as “...the 
process of increasing participation in international operations.” (Welch & Luostarinen, 1988, p. 
36). The above term takes both inward and outward growth of foreign firms into account. The 
expansion of countertrade, for example, in the form of barter or buy-back arrangements, clearly 
shows how outward growth is related to inward growth (Huszagh & Huszagh, 1986; Khoury, 
1984). The merits of Welch and Luostarinen’s (1988) definition have been recognised by others 
(Young, 1990) because it is a working explanation that is concise and readily interpretable. 
Furthermore, it is comprehensive enough to account for the many variables associated with 
international expansion. 
 
Moreover, the internationalisation process of firms has been subject to widespread research 
attention and empirical investigation (Anderson, 1993). Welch and Luostarinen (1988) have 
comprehensively reviewed this literature and concluded that “…there is a wide range of potential 
paths any firm might take in internationalisation” (p. 43). Various approaches and viewpoints have 
contributed to today's perception of firm internationalisation. To help clarify the systemic and 
21 
 
University of Ghana http://ugspace.ug.edu.gh 
behavioural issues underlying internationalisation theory, economic, econometric, organisational, 
marketing, and managerial models, for example, have been developed (Dalli, 1994). Nonetheless, 
amid such a flurry of interest from such scholars, one approach has generated a substantial body 
of literature on the topic of internationalisation: exports development. 
 
Given that the firm's export activities are constantly changing, the mode of export behaviour 
appears to be production (Albaum et al., 1989) and a significant number of studies have used this 
frame of reference to investigate the complex and evolutionary mechanism of internationalisation 
of firms. It has been argued that the literature on internationalisation, from the standpoint of export 
growth, is one of the most advanced and mature areas of expertise in international business (Haar 
& Ortiz-Buonafina, 1995).  
 
2.2.7 Effects of International Trade 
This section reviews the literature on how revenue and tax capacity affect international trade and 
economic growth and the effects of exports and imports on economic growth.  
 
2.2.7.1 How Revenue and Tax Capacity affect International Trade 
Regarding revenue and taxes, and data indicates that the amount of government revenue collected 
is determined by the taxation capacity of individual countries, the taxation goals set by the 
authorities, and the willingness of governments to raise revenue. However, the effectiveness of 
maximising revenue capacity and meeting taxation goals is dependent on a variety of other factors. 
These factors include the macroeconomic climate, economic structure, and level of growth, as well 
as administrative ability and tax willingness (Teera & Hudson, 2004). These factors can interact 
22 
 
University of Ghana http://ugspace.ug.edu.gh 
in various ways at different times and in different countries, resulting in differences in government 
revenue collection across countries. 
 
Inflation, public debt, and foreign aid are all macroeconomic variables that influence government 
revenue mobilisation. Inflation in this context is a surrogate measure for the consistency and 
stability of a country’s macroeconomic policies. This reflects its direct impact on tax collection 
through its effects on consumption, expenditure, and other tax categories (Davoodi & Grigorian, 
2007). Higher inflation rates demoralise the public, decreasing tax enforcement and reducing 
revenue collection (McMahon & Schmidt-Hebbel, 2000). In general, inflation harms government 
revenue mobilisation.  
 
Furthermore, foreign aid, whether in the form of grants or loans, has an effect on the country's tax 
effort. The response of the tax effort to increased aid flows can be positive, negative, or zero, 
depending on the intent of the aid and whether it is a grant or a loan. Since governments are not 
required to repay grants, they have a negative relationship. Grants eliminate incentives for 
governments to implement good policies and retain effective structures, resulting in low tax 
revenues due to tax breaks for influential interest groups and poor tax enforcement, as well as 
diverting resources away from fixing governance flaws. In theory, a loan that a government 
promises to repay will induce the government to raise revenue to service the loan (Gupta et al., 
2003; Agbeyegbe et al., 2006). Furthermore, the instability of foreign aid has consequences for 
macroeconomic stability, which can impact the country's tax effort (Gupta, et al. 2003). 
 
23 
 
University of Ghana http://ugspace.ug.edu.gh 
Public debt also plays a role in determining the extent to which countries may take advantage of 
their taxable capacity (Tanzi, 1987). The influence of government debt is determined by its scale 
and how it is funded. A high public debt necessitates the government raising taxes to service the 
debt (Teera & Hudson, 2004). As a result, a high public debt has a favorable relationship with 
existing and future tax levels. However, if the majority of the debt is external, debt servicing causes 
macroeconomic imbalances that may result in a reduction in taxation. This is due to the fact that 
servicing international debt necessitates a trade account surplus, which necessitates a reduction in 
imports. 
 
The economic structure of a country also explains its tax power. The proportions of agriculture 
and industry in GDP, population size (such as population density and dependency ratio), and 
urbanisation all represent the economic structure of the country. Agriculture's GDP contribution 
in developing countries tends to be insignificant due to the general difficulty of taxing agriculture 
and the government's deliberate provision of tax breaks and/or subsidies to the industry. As a result, 
the agricultural sector's share is inversely linked to government revenue mobilisation. This is due, 
in part, to the fact that the agricultural sector in most LDCs is dominated by small farmers, who 
are notoriously difficult to tax, particularly when a large portion of agriculture is subsistence.  
 
Moreover, most agricultural activities organised in small-scale farming generate limited taxable 
surpluses (Davoodi & Grigorian, 2007; Suliman, 2005; Agbeyegbe et al., 2006; Khattry & Rao, 
2002; Teera & Hudson, 2004)). In general, as agriculture's share of GDP grows, less revenue is 
received. However, in countries where agricultural products account for a significant proportion 
of total exports, the relationship may be positive (Agbeyegbe et al., 2006). 
24 
 
University of Ghana http://ugspace.ug.edu.gh 
Industrial operations, on the other hand, are easier to tax in most countries because company 
owners keep better records and their activities are concentrated in cities. Unlike the agricultural 
sector, the manufacturing industry will produce greater surpluses, which benefits the tax base 
(Teera & Hudson, 2004; Tosun & Abizadeh, 2005). The share of the mining sector is also said to 
exhibit a positive relationship with tax revenue (Agbeyegbe et al., 2006) however, when resources 
are correlated with a higher probability of conflict, this relationship may be negative (Agbeyegbe 
et al., 2006; Ndikumana, 2004). 
 
Population density is also an important factor that controls for the scale effect of the economy. 
According to economic intuition, if there are economies of scale in tax collection due to fixed 
administrative costs, then the higher the population density, the higher the taxes raised for a given 
tax base and tax ratio (Khattry &  Rao, 2002; Teera &  Hudson, 2004). 
 
2.2.7.2 Exports and Economic Growth 
Causality studies yield a wide range of findings about the relationship between exports and 
economic development (Giles and Williams, (2000); Gries & Redlin (2020)). It is possible to claim 
that there is no universal agreement on causality and its course in all countries. Indeed, while some 
researchers conclude that export has a single causal relationship with economic growth, implying 
that export promotes economic growth (Chow, 1987), others discovered that export has a single 
causal relationship with economic growth, implying that causality runs from economic growth to 
export growth (Giles and Williams, 2000). Furthermore, some argue that there is a two-way causal 
relationship between economic development and export, while others argue that there is none. It 
can be seen from this that there are primarily four findings about the causality between export and 
25 
 
University of Ghana http://ugspace.ug.edu.gh 
economic growth based on a large number of country cases. The first is that the chain of causality 
extends from export to economic development. The second conclusion is that causality runs from 
economic growth to export. At the same time, the third and fourth argue that there is two-way 
causality between economic growth and export and that there is no causal link between the two. 
This demonstrates how broad the theoretical background of causality in this problem is (Chen, 
2009). 
 
However, the fundamentals of the export-led growth hypothesis indicate that increasing aggregate 
exports positively affects economic growth in two ways. Second, aggregate exports add to the 
country's circular flow of income by rising production levels through the multiplier impact. 
Furthermore, the level of exports raises foreign exchange receipts, leading to imports of critical 
inputs into the production system (Öztürk & Acaravcı, 2010). If the factors that affect export are 
considered, it could be argued that a country's degree of foreign openness in terms of rising 
competitiveness in domestic markets by increasing productivity and decreasing costs will enable 
it to increase its exports. Adopting advanced technologies, managerial expertise, and in-depth 
discovery of foreign markets can also be seen as export-boosting factors. Significantly, one factor 
that encourages exportation is ensuring adequate capital accumulation in a region, which could be 
provided by growing capital inflows into the country. As a result, the country is able to increase 
production with new capital and expand exports with rising economic growth (Temiz & Gökmen 
2010). 
 
Even though the relationship between foreign trade and economic growth is studied using either 
export-led growth or growth-led export approaches, the foregoing claims highlight the role of 
26 
 
University of Ghana http://ugspace.ug.edu.gh 
importation in the process of exportation and economic growth. As a result, increasing imports 
may be viewed as a complement to exports. By relying on technology-intensive intermediate 
products, a country can maintain productivity by fostering the processes of research and 
development, learning, and technology adaptation. As a result, imports of these intermediate 
products could serve as the foundation for the production of goods for export, implying a kind of 
complementarity between imports and exports (Kotil & Konur, 2010). 
 
2.3 TRADE MIS-INVOICING/SOURCES AND TYPES OF TRADE MIS-INVOICING 
There is convergence in the literature regarding the definition of trade mis-invoicing. According 
to Salomon (2018), trade mis-invoicing refers to voluntary falsification by at least one party to the 
international transaction of the quantity and/or classification of the goods and services. Similarly, 
Choi and McGauran (2018) argue that trade mis-invoicing is characterised as dishonest cases in 
which both importers and exporters misuse or in their customs declarations handle the value (e.g. 
price, sum or quality) of trading goods. It is a means of moving money illegally across national 
boundaries. Trade mis-invoicing is typically not observed directly. A negative trade deficit does 
not automatically mean that there is trade mis-invoicing.  A trade gap on its own cannot identify a 
person responsible. Trade mis-invoicing takes place if the true value of imports and exports differ 
from the number of exports or imports that the authorities report. 
 
There are a number of ways trade mis-invoicing could occur. However, four (4) broad categories 
have been highlighted in the literature. These four common trade mis-invoicing categories are 
import under-invoicing, over-invoicing imports, export under-invoicing and export over-
invoicing. Every subcategory of trade mis-invoicing gives the parties concerned many advantages. 
27 
 
University of Ghana http://ugspace.ug.edu.gh 
The undervaluation of exports includes under-reporting the export amount from a nation to escape 
taxation on corporate income in the importing country by depositing the difference in value into a 
foreign account. Similarly, export over-invoicing means reaching the export amount leaving a 
country that also helps the vendor obtain additional export credits. Companies and individuals may 
also use this form of trade mis-invoicing to conceal capital flows, to prevent capital controls or 
scrutiny against the use of money laundering.  
 
On the import side, traders also under-report the import amounts to avoid applicable import tariffs 
and VATs, which may occur due to the tariff and tax regime on the products. If an importer over-
reports its imports, they frequently legitimise the transfer of additional capital under legitimate 
export payments. The over-invoicing of imports masks capital movements from a region. That 
could be an asset for capital controls, and a corporation could deduct the amount of production 
from its report on year-end sales to the government, lowering the taxes it owes to the government.  
 
2.3.1 The Arm’s Length Principle 
Since 1979, the OECD has been developing practical guidance to implement the arm’s length 
principle. The arm’s length principle is the method recommended by the OECD, which in its 
Model Tax Convention on Income and Capital, adopted the arm’s length standard in article 7, 
related to the business profit, and article 9, on associated enterprises (OECD, 2001). The OECD 
Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (thereinafter 
OECD TPG) are continuously revised and updated with new guidance to deal and manage with 
the enormous changes and challenges posed by an increasingly globalised economy. This principle 
has been the subject of five OECD reports in 1979, 1984, 1995 and 2010 (OECD, 2001). The last 
28 
 
University of Ghana http://ugspace.ug.edu.gh 
update particularly took account of the revision of the guidance on comparability and profit 
methods and the new guidance on the transfer pricing aspects of business restructurings. 
 
Additionally, the OECD TPG, released in 1995, was the culmination of the efforts made by the 
international tax community to review the existing standard and adapt them to the modern business 
world. The OECD Transfer Pricing Guidelines 27 made it clear that transfer pricing should not be 
confused with tax fraud or tax avoidance, even though transfer pricing transactions may be used 
for these purposes (Markam, 2004). Also, since transfer pricing may have other purposes aside 
from tax avoidance, fiscal authorities should not automatically determine that companies with 
cross-border activity are trying to manipulate profits. Mainly because it is not easy to establish the 
price market in a precise way. In this context, the Fiscal Affairs Committee of the OECD created 
a set of rules to reduce the risk of misunderstanding or abuse concerning the taxation of some 
operations in groups of companies by adopting the so-called arm’s length principle. 
 
The arm’s length principle is usually applied by comparing the ‘conditions’ (e.g. price or margin) 
of a controlled transaction with those of independent transactions. The OECD has set forth a series 
of accepted methodologies according to Transfer Pricing Guidelines (1995). They are the 
comparable uncontrolled price (hereafter CUP) method, cost-plus (hereafter CP) method, resale 
price method, profit split method, and transactional net margin method. Among these methods, the 
CUP and the CP methods are the two most frequently used regulation rules (Ernst & Young, 2010). 
Again, the arm’s length principle is usually translated into marginal cost pricing in the literature 
(Kind, Midelfart, & Schjelderup, 2005; Peralta, Wauthy, & van Ypersele, 2006). This is due to the 
fact that, in order to detect deviations from pricing in practice, tax authorities compare data from 
29 
 
University of Ghana http://ugspace.ug.edu.gh 
the audited firm to data from comparable transactions between independent buying and selling 
firms. The general idea is that as long as the audited firm cannot influence the comparable 
transactions, then the data from either the CUP or the CP method should approximate the 
independent price of a competitive market. This situation implies that the arm’s length principle 
can be proxied by marginal cost pricing in a competitive market. 
 
Furthermore, the OECD notes in its 1995 study that the key justification for promoting the arm's 
length principle is to encourage the growth of investment and foreign trade because the principle 
considers related parties as being in similar tax positions as separate parties, thereby ensuring 
competitiveness. Furthermore, the OECD 1995 study concluded that the arm's length principle 
successfully resolved the vast majority of transfer pricing situations (OECD, 2001). However, even 
though the principle can be viewed as simple, its application has proved difficult (Elliot & 
Emmanuel, 2002). 
 
In practice, applying the arm’s length price has two main weaknesses. Firstly, it is difficult to find 
equivalent transactions that are the same or identical to evaluate the arm's length price or market 
price, mainly when dealing with intangibles. Second, various taxpayers and tax authorities can 
have different conceptions of the best method for determining the arm's length price since different 
approaches may result in substantially other arm's length prices. The United States transfer pricing 
guidelines use a range method to define the arm's length price when dealing with this issue. As a 
result, if the measured price or margin falls within the range of arm's length prices, tax authorities 
need no adjustment. The arm's length theory is based on the comparability of the same transaction 
between unrelated parties in the same situation. However, finding this level of sameness condition' 
30 
 
University of Ghana http://ugspace.ug.edu.gh 
is problematic. As a result, for practical purposes, the theory is interpreted as ‘comparable 
transactions' between unrelated parties in ‘comparable circumstances.' (Feinschreiber, 2004).  
 
Several tax scholars such as Owens, Hamaekers and Avi-Yonah believe that the arm’s length 
principle is still relevant and should be used when determining the arm’s length price (Owens, 
2005). They argue that the continued presence of the arm's length principle since its 
implementation more than 70 years ago demonstrates that the principle is adaptable enough to 
adapt and evolve with current economic situations. This, they say, will continue in the future. 
 
On the other hand, opponents of the arm's length theory contend that it would be impossible to 
apply in the future because multinational corporations' businesses are so intertwined. They claim 
this may cause problems under the arm's length principle because MNEs profit from combining 
some of their corporate functions to decide the price sold from production to sales divisions. Again, 
maintaining bookkeeping records of corporations' transactions with their associates in- the country 
would be costly. Another concern related to the arm's length approach's shortcomings is that it 
would be difficult to apply to cross-border electronic commerce because the OECD Transfer 
Pricing Guidelines favour conventional transaction-based transfer pricing methodologies 
(Hellerstein, 1975). In general, global formulary apportionment measures an MNE's total income 
as a company and divides the benefit among its subsidiaries based on specific formulae (Weiner, 
2002). Electronic commerce transactions, for example, are more challenging to track than non-
electronic transactions. This is due to the ease with which the server used in electronic commerce 
transactions can be transferred from one location to another. As a consequence, locating 
comparable data for such a transaction would be difficult. These issues pose significant challenges, 
31 
 
University of Ghana http://ugspace.ug.edu.gh 
especially in developed countries with the limited administrative capacity to track transfer prices 
between associated parties (McLure, 2002). 
 
Many scholars criticise its use as a substitute for the arm's length principle at the global 
(international) level, claiming that replacing the arm's length principle with global formulary 
apportionment is not a good solution to transfer pricing issues (Amerkhail, 2002; Engel, 1998; 
Neilsen et al., 2001).  Furthermore, Nielsen et al. (2001) study reveal that the formulary 
apportionment method does not remove the profit-shifting problem but may increase the practice 
of profit-shifting by MNEs in oligopolistic competition.  Another study by Devine, O’Clock and 
Seaton (2006) shows the potential for double taxation due to the implementation of different 
formulary apportionment models. 
 
As a result, the arm's length price is the only concept generally recognised by most countries 
worldwide in deciding the transaction price between related companies. Supporters of the arm's 
length approach believe that replacing the arm's length principle would jeopardise international 
consensus among countries, potentially leading to double taxation (Roin, 2008).  The OECD 
Transfer Pricing Guidelines and the 1979 OECD Report reject using global formulary 
apportionment to replace the arm’s length principle. However, recent developments show that the 
OECD Transfer Pricing Guidelines’ stand on this matter has softened. In its draft revision of the 
Guidelines, the OECD compromised using formulary apportionment to allocate profits (Bell, 
2009). 
 
32 
 
University of Ghana http://ugspace.ug.edu.gh 
Furthermore, in 2007, the European Commission approved a five-year pilot project to investigate 
the use of formulary apportionment within the European Union for small and medium-sized 
enterprises (Tropin, 2007). Though the OECD guidelines include a variety of methodologies for 
determining an arm's length price, each method's implementation may result in different arm's 
length prices. The OECD's preferred approaches, known as traditional methods, focus on similar 
data from third parties. As a result, MNEs can have the advantage of being able to choose one 
form over another. 
 
2.4 MEASURING TRADE MIS-INVOICING 
It is impossible to directly measure the magnitude of trade mis-invoicing because they conceal in 
countless trade transactions. Empirical studies have estimated trade mis-invoicing by comparing 
trade discrepancies using the PCTG method or using the Price Filter Method (World Customs 
Organization, 2018).   
 
2.4.1 Trade Gaps/Trade Discrepancy 
The PCTG method of measuring trade mis-invoicing is based on the concept of trade gaps/trade 
discrepancy. Further, Choi and McGauran (2018) define trade discrepancy as the difference 
between the importing country's trade value and the exporting country's value. In fact, trade 
disparities at the most divided level (i.e. transactional level) may indicate trade mis-invoicing. 
However, governments are hesitant to exchange data with trading partners due to the sensitive 
nature of transaction-level data. As an alternative, aggregated trade data from foreign 
organizations, grouped by the Harmonized System (HS) code and partner countries, has been used 
to study disparities in trade. However, it should be noted that the variations in commerce could 
arise from commercial data collection due to statistical and logistical reasons. Accurate trade 
33 
 
University of Ghana http://ugspace.ug.edu.gh 
mismanagement interventions can and must aid customs authorities' work to curb illicit trading. 
Export underestimations intend to mask international exchange income, i.e. tax havens.  The logic 
is that if a trade transaction has an abnormal unit price and irreconcilable differences in the trade 
value ascribed by the trading partner, it would be reasonable to suspect under-valuation or over-
valuation. 
 
 The concept at the core of the PCTG method is the trade gap. The value of a particular 
international commercial transaction is described in the discrepancy (if any) separately stated by 
the two parties. Legitimate reasons for trade discrepancies include the following: the Cost 
Insurance Freight (CIF) and Free On Board (FOB) ratio, differences between trading partners 
regarding classification in the harmonized system, attribution of trade partners, foreign exchange 
rates and timing and low-value thresholds. 
 
2.4.2 The Partner Country Trade Gap (PCTG) Method 
The PCTG method was developed by Bhagwati (1964a). It uses mirror statistics whereby one 
country's exports are another country's imports and should balance out if there is no mis-invoicing. 
Usually, a 10% filter is allowed to cater for transportation costs. GFI updated this markup from 
10% to 6% following updates to the OECD International Transport and Insurance Cost (ITIC) 
database. (Collin 2020). The PCTG has been widely used to estimate mis-invoicing, especially by 
institutions such as Global Financial Integrity (GFI).  
 
2.4.2.1 Limitations of the PCTG Method 
There have been several criticisms of the PCTG approach.  These have been summarized by GFI 
(2017) as follows:  
34 
 
University of Ghana http://ugspace.ug.edu.gh 
 
Incorrect to assume that trade statistics in advanced economies exhibit no mis-invoicing: One 
fundamental premise of partner country trade statistics is that they consider commercial statistics 
of developed economies as arms-length values for comparing commercial statistics of developing 
countries. However, a study by Hong and Pak (2017), using aggregate trade statistics from the 
IMF DOTS database and transaction-level trade data from customs agencies, show that a 
significant degree of trade mis-invoicing occurs in advanced economies. 
 
Unobserved trade costs: Typically, import transactions are valued at a Cost, Insurance and 
Freight (CIF) basis. This has to be converted to Free-On-Board (FOB) format used for exports 
before any trade gaps can be calculated. Unfortunately, because data are usually not reported on 
transaction costs of trade, adjustments of 10% rule-of-thumb for these transactions are made. The 
thumb rule is built from a regional export-import gap. It means that a 10% minimum trade cost 
calculation cannot accurately convert bilateral trade values into all goods, with varying costs of 
transportation and insurance. 
 
Use of aggregate trade statistics: The literature in this area concentrated on calculating trade 
deficiencies, using total annual exports and import figures. For some instances, however, bilateral 
data may not be available between trading partners for all exchanged goods and products, 
contributing to the creation of highly misleading trade loopholes estimates. However, using 
aggregate trade loopholes can mask or over-invoice particular goods that cancel the aggregate. 
 
35 
 
University of Ghana http://ugspace.ug.edu.gh 
Data unavailability for certain countries, years, or commodities: A comprehensive collection 
of bilateral trade figures are usually not applicable to all global trading nations for all goods. This 
can lead to misleading estimates of trade gaps being calculated. For instance, there has been 
missing data on cocoa exports from Ghana for several years in the International UN COMTRADE 
database. 
 
Exports and import transactions can be recorded in different years: International trade will 
take considerable time to complete according to the mode of transport and the distance between 
trading points. This could lead to the business partners recording the same transaction over various 
years, while trade gaps are calculated for one year using statistics.  
 
Entrepôt trade: For several goods, the countries of origin and destination reported are reports 
from middle ports, where transportation is processed until it is eventually transported to its final 
destination. Furthermore, foreign commodity trading firms may agree to redirect transportation to 
warehouses to take advantage of arbitrage opportunities emerging when price rates fluctuate. 
Double accounting is given when transit countries and exporting-importing partners record export 
and imports that entrepôt ports. For example, some exporters from neighbouring countries, such 
as Burkina Faso, bring their gold to Ghana for assaying before they export. 
 
Exchange rates used for currency conversion: International traffic transactions occur for certain 
vehicle currencies like, the US dollar or local currencies. Consequently, if trading partners use 
different exchange rates to convert trading values into USD, their trading statistics may have gaps. 
36 
 
University of Ghana http://ugspace.ug.edu.gh 
However, some developed countries maintain multiple exchange-rate regimes that increase the 
risk of such errors. 
 
Country idiosyncrasies: Countries that agree not to disclose bilateral trade flows for particular 
goods for specific years may be subject to an estimation of market gaps. Although Switzerland, 
for example, is a major importer of gold, its international trade in unmanned gold was not recorded 
until 2012. Also, there are some missing statistics for Ghana's exports in the UN COMTRADE 
database. 
 
Mis-invoicing, as measured from macroeconomic data, appears to represent a specific cancellation 
between companies who may be faulted (over-invoicing and under-invoicing). Therefore, the use 
of aggregate statistics is likely to underestimate the true size of gross capital flow by mis-invoicing 
(Patnaik, Gupta and Shah 2012). 
 
2.4.3 Price Filter Methods (free-market price filter and interquartile range price filter) 
For each Harmonized Commodity Description and Code System (HS code), the price filter 
methods estimate price filters (price filtering process) as a proxy for prices at arm's length and use 
price filters to identify suspicious transactions with odd prices, a sign of possible trade mis-
invoicing. Trade mis-invoicing occurs when a reported transaction's unit price differs from the 
transaction's arm-length price. The arm's length purchase price can vary depending on the specific 
situation of the deal, including contractual conditions, economic circumstances, and the company's 
business strategies. The method of price filtering is mainly used to estimate trade defaults.  The 
price filter approach can effectively and at a low cost detect unusually priced transactions without 
37 
 
University of Ghana http://ugspace.ug.edu.gh 
using partner transaction data. This can be used routinely and effectively in computerised real-
time tracking of transactions and historical records.  
 
There are two primary methodologies under the price filter analysis.  These are the:  inter-
quartile range filter and the free-market price filter. 
 
2.4.3.1 Free-market price filter 
The free-market price filter method compares real unit price levels for a particular commodity with 
its contemporary free-market price, plus/minus an appropriate price volatility filter, particular trade 
contracts between firms, commodity types, and various countries' issues. Transactions that vary 
significantly from bad prices beyond what is explained by fair market fluctuations are either 
overvalued or undervalued.  
 
The free-market price filter method has one advantage: researchers do not quantify the low price, 
as the free-market price is readily available.  Also, the free-market price filters have some other 
advantages over the interquartile price filters, such as the following. Unlike the interquartile price 
filter, the free-market price filter accounts for variation in monthly commodity prices.  
 
The free-market price filter is measured separately from the declared import values, while the 
prices for inter-quartiles dependent on declared import values are evaluated. Therefore, related 
party transaction information does not influence the free-market price filter in the import and 
export database. However, to apply this approach, a free-market price is needed that cannot be 
readily available, mainly if the commodity market is not developed (Hong et al., 2014).  
 
38 
 
University of Ghana http://ugspace.ug.edu.gh 
2.4.3.2 Interquartile Range Price Filter 
The inter-quartile price filter assumes that unit prices within the 25th and 75th percentile of the 
observed distribution of unit prices for a specific commodity denote the arm's length price range. 
Any transaction that falls above or below this price range is categorized as overvalued or 
undervalued. Thus, the overvalued transactions are identified by the values declared above the 
upper-quartile price range whilst the undervalued are those declared below the lower quartile price 
range.  
 
2.4.3.3 Limitations of the Price Filter Method 
According to Reuter (2011) and Carbonnier and Zweynert de Cadena (2015), the study of the price 
filter based on commercial micro-data is usually intuitive. However, it is still subject to the 
following methodological limitations: 
 
Endogeneity of the chosen price filter: The lower and upper limits are calculated based on 
observed trade transactions to determine the interquartile price range. The 25%, and 75% 
percentiles use implies that 50% of the transactions are abnormally priced. Therefore, as long as 
each commodity's prices are at least somewhat variable, it will still generate overpriced and 
underpriced transactions. 
 
Sensitivity to outliers: Because price filters are often calculated using the whole price distribution, 
a reclassification of transactions from usual to abnormally priced occur as a result of statistical 
outliers, possibly due to human recording errors and vice versa.  
 
39 
 
University of Ghana http://ugspace.ug.edu.gh 
Potentially disproportionate impact of related party transactions: The price filter analyzes 
cannot differentiate between transactions of related or unrelated entities without clear details on 
the trading companies. It means that where related parties control foreign trade with an opportunity 
for deviating from prices at arm's length, the projected interquartile range may be prejudicial. 
 
Product heterogeneity: Price filter analysis typically relies on a high disaggregation level product 
classification, generally when it is differentiated by the 10-digit HS code point. However, this 
approach can also wrongly classify high-end products as overpriced, low-end products underrated 
for goods that are rather heterogeneous concerning quality and prices. In the meantime, costly mid-
range transactions may be wrongly listed as legal transactions. 
 
Limitations in product classification system: Whilst most traded goods are described explicitly 
under the HS scheme, there are also product codes that collect different kinds of products that 
conform to the current classification. The product classes will produce an indeterminate amount 
of statistical noise in estimate price filters through a standard study of multiple product 
subcategories. For example, a catch-all other subcategory is included in a 2009 harmonized tariff 
schedule 12,581 out of 28,985 product categories. 
 
Quantity faking: Price filter systems through several goods frequently fail to recognise the 
potential mis-invoicing of amounts instead of values, i.e. the reporting of exchange quantities or 
the reporting of them. Given that unit prices for transactions are a basis for price filter analysis, 
the estimates of the exchange market misprice may often vary from or underreport the quantities. 
 
40 
 
University of Ghana http://ugspace.ug.edu.gh 
Large quantity transactions with small price differences: Large quantitative transactions in 
which reported prices only vary by a small margin because prices for the product will stay within 
the price range of the arm length, whereas the overall mispriced sum may still be significant. 
 
Similarly, Pak (2018) also identified some limitations of the price filter methods as follows: 
product heterogeneity, different contract information, such as long-term contracts vs short-term or 
spot transactions, endogeneity in statistical price filter estimation (interquartile price range filter), 
clerical or tracking errors in data, missing identification of transactions with a large quantity but 
mispriced by a small margin, and goods with unpredictable price movements. Therefore, since not 
all the transactions that are priced abnormally are mispriced, customs may need to check 
commercial records concerning all the suspicious transactions found by price filters. The estimated 
sum should be treated with reservations when a pricing filter method is used to estimate mis-
invoicing amounts because the price filter method may contain false positives and fail to recognise 
faulty transactions. 
 
Empirically, several studies have used the price filter methodology focusing mainly on the United 
States due to data availability (Pak et al., 2003; de Boyrie et al., 2005; Hong and Pak, 2017). These 
studies have examined the magnitude of trade mispricing between America and its trading partners 
using pricing filter mechanisms to estimate millions of import-export transactions. Zdanowicz et 
al. (1999) studied examined commerce between Brazil and America to document the quantity of 
capital flight covered by commerce in mispriced goods. Their findings show that capital flight 
from Brazil to the USA through a trade mispricing took place in a single year, 1995, of 2-4 billion 
USD.  
41 
 
University of Ghana http://ugspace.ug.edu.gh 
Pak et al. (2003) used a similar approach in the Greece-US exchange mispricing study. The authors 
consider that mispriced transactions have moved from Greece to the United States between USD 
132-276 million and from Greece to the other part of the world in 1995 to about USD 5.5 billion. 
de Boyrie et al. (2005) studying the trade data from 1995-2000 between Switzerland and the USA, 
revealed that the economic mispricing of USD31 billion was used to transfer the Swiss to the 
United States. They also analysed the trade between Russia and the United States in imports and 
exports from 1995-99 to find 7.24 billion US dollars on undercutting Russian exports and 1.68 
billion US dollars on over-invoicing Russian imports. 
 
Hong et al. (2014), used interquartile price filters, arm's length price filters, and partner country-
mirror data to derive comparative estimates of trade misprice in banana trading with Latin 
American and Caribbean countries. The authors conclude that although there is little evidence of 
substantial errors in inter-quartile and partner-country trade studies, the use of free-market, low-
cost pricing filters indicated the existence of significant mispricing. The results show that 
underestimated transactions constitute 54 per cent of US importers' total banana imports reported 
in 2000-2009. 
 
2.5 DETERMINANTS OF TRADE MIS-INVOICING 
It is essential to understand and investigate the determinants of trade mis-invoicing. This is because 
it is crucial to understand the factors influencing trade mis-invoicing since the resources lost 
through the trade mis-invoicing could have been used to significantly contribute to economic 
development. Trade mis-invoicing is a loss of money as these resources leave the country without 
domestic control, thus reducing the funds required for growth and development. (Yalta & Demir, 
42 
 
University of Ghana http://ugspace.ug.edu.gh 
2010). Trade mis-invoicing often adversely affects income distribution among many people as a 
result of moving funds from the jurisdiction of fund creation, escaping relatively high taxes. This 
inevitably erodes the national fiscal base and reduces public income (Farzanegan, 2009).  
 
There has been exciting literature researching the reasons for trade mis-invoicing by examining 
the data on bilateral trading partners (for example, Fisman and Wei, 2007), Berger and Nitsch 
(2008) and Farzanegan (2008). Trade mis-invoicing usually stems from financial benefits, such as 
gaining a black currency premium or avoiding taxes and tariffs (Buehn & Eichler, 2006).  
 
Several drivers can be grouped under headings such as political, legal and regulatory, economic 
and financial. Economic and financial drivers may include corporate income tax rates, tax rate 
differences, indebtedness, Import-Export taxes and duties, capital controls, interest rates and 
differentials in interest rates, domestic inflation, real effective exchange rate overvaluation 
(REER), capital account openness, exchange rate regimes, macroeconomic instability, large 
budget deficits, low growth rates, the spread between foreign and domestic interest rates and real 
exchange rates. Some political drivers of mispricing include patronage, political freedom, 
accountability, political instability, corruption and public-sector reforms. In contrast, some legal 
and regulatory drivers are tax disclosure requirements, transparency initiatives and industry-
specific regulations. 
 
The most common cause of illicit financial flows are corruption, fraudulent funds usage and tax 
evasion. These are therefore mainly a question of governance resulting from weak institutions and 
insufficient regulatory structures. These include limited access to financial and legal expertise and 
43 
 
University of Ghana http://ugspace.ug.edu.gh 
misaligned policies on domestic tax and commercial reporting with global requirements. IFFs also 
thrive in the presence of obsolete knowledge gathering and communication systems and a lack of 
political will. A business which wishes to transfer capital from a country will underfund its exports, 
bringing reduced exchange to the government (Patnaik et al., 2012). 
 
One political factor that drives mispricing is corruption. Berger and Nitsch (2012) study the 
relationship between trade mis-invoicing/mispricing and corruption from 2002 to 2006 for the 
world's top five importers: the US, Germany, China, the UK and Japan. They conclude that trade 
mis-invoicing by the partner countries increases with the level of corruption. Corruption would 
potentially lead to capital flowing out of Canada, distorting the economic and financial systems by 
enabling individuals to take control by patronage rather than capacity and the prevalence of 
approved bribes. 
 
According to Pak (2014), trade mis-invoicing is driven by income tax avoidance or evasion, 
overvalued domestic currency, domestic inflation, import duty avoidance, export incentives, 
capital flight and money laundering.  As IFFs are typically caused by corruption, illegality and tax 
evasion, they are considered a governance problem resulting from weak institutions and inadequate 
regulatory infrastructure. These include limited access to financial and legal expertise and 
misaligned domestic tax and commercial reporting policies with global requirements. They also 
have outdated systems to collect and exchange information and a lack of political will. 
 
The current literature indicates that poor financial, administrative and legal institutions can abuse 
discretionary powers (see Bardhan 1997; Andvig et al. 2001; Aidt 2003). Under this situation, IFFs 
44 
 
University of Ghana http://ugspace.ug.edu.gh 
arise from the will to hide illegally accumulated wealth abroad and not necessarily because of 
interest rates or macro-economic distortions (Kar and Cartwright-Smith 2008, 2010; Heggstad and 
Fjeldstad, 2010) 
 
Traditionally, high customs duties (de Boyrie et al. 2007; Boyce and Ndikumana 2001; Beja et al. 
2005) is given as a reason for trade mis-invoicing. Patnaik et al. (2012) argue that businesses can 
understate the actual value of goods while paying higher customs rates or VAT on goods or are 
subject to quantitative restrictions. To get reduced currency into the country, an enterprise which 
wants to transfer capital out of the country will under-report its exports. The evidence supports the 
argument that the strongly linked situation of foreign debt and capital flight occurs due to 
privileged classes taking money from their countries and dissimulating the profits abroad 
(Bracking 2003). 
 
Some political drivers of mis-invoicing include patronage, political freedom, accountability, 
political instability, corruption and public-sector reforms. In contrast, some legal and regulatory 
drivers are tax disclosure requirements, transparency initiatives and industry-specific regulations.  
 
The root cause of all IFFs is a lack of political will, not inherently bad management practices or 
capacity limitations. The absence of a political desire to create a legal and efficient state is thriving 
in tax evasion, mis-invoicing, transfer rates, corruption, and trafficking in persons and drugs. Such 
effectiveness, including tax systems efficiency, stems from formal and informal institutionalised 
arrangements or policy settlements that establish the legitimacy of government, encourage 
prosperity and increase public income. The undertakings of such arrangements is unconstitutional 
45 
 
University of Ghana http://ugspace.ug.edu.gh 
and illegal. Political leaders and stakeholders must understand that efficient taxation structures are 
required to provide the state with the necessary resources to ensure that its property rights are 
protected and that political stability and economic development are fostered. The degree and form 
of tax avoidance result from a political consensus that taxes and enhances the institutional capacity 
to do so effectively. This shapes and demonstrates taxpayers' inherent readiness (tax morale) to 
pay taxes.  
 
Patnaik et al. (2012) finds that in countries with significant commercial flows, capital controls are 
associated with high trade mis-invoicing. They find that faults are a channel of openness to de 
facto capital accounts. Capitals switch to international shores in countries with high current-
account deficits. A continued current-account deficit manifests economic instability and causes 
owners of capital to transfer resources to foreign regions. To boost the current account balance, a 
nation that faces constant current account deficits will take on a devaluation.  This can be achieved 
either by a direct appropriation/nationalisation of private properties or by inflationary taxation. 
Alternatively, it can internally raise capital by facilitating a private sector switch. In any case, the 
private sector would be allowed to withdraw its assets from the government's control. 
 
According to some studies (Quereshi and Mahmood 2016; Patnaik et al. 2012), some traders are 
encouraged by high customs duties to declare their goods a lower value to avoid payment higher 
tariffs. Highly indebted countries will probably experience more significant capital outflow. 
Several studies focusing on the African countries, including Khan and Ul-Haque (1985), have 
shown that the government typically borrows from foreign countries and multilateral agencies, 
while the private sector shifts money from foreign countries to other countries. Also, disposing of 
46 
 
University of Ghana http://ugspace.ug.edu.gh 
foreign exchange resources by capital flight creates more demand for foreign debt. Khan and Ul-
Haque (1985) also point out that debt countries' residents are expropriated, implying that they can 
confiscate the government's domestic assets through open nationalisation, taxes, or exchange 
checks. In contrast, the risks of the like assets held abroad are negligible. Indebtedness is the ratio 
from the WDI database of external debt to GDP. 
 
Yalta and Demir (2010) examine the extent of trade mis-invoicing for 1970–2007, surveying 
Turkish exports to its major trading partners. They conclude that exports are not paid, although 
Chinese imports are over-filled. They also examine the effects on commercial countdowns of 
customs unions and trade liberalisation policies and conclude that liberalisation policies hurt 
imports' overall price. 
 
Mahmood (1997) examined data collected for 96 products and imports of six industrialised 
countries. The significant determinants of import under-invoicing in Pakistan during 1981-88 are 
investigated. The impacts of import taxes and nontariff import controls are tested: import taxes are 
perceived to be the highest and have the most significant component with a solid connection to 
import under-invoices. The rate variable is very relevant and optimistic for import under-invoicing. 
This is in line with Mahmood (1997) and Patnaik et al. (2012), who find the tariff rate to be highly 
and positively associated with import under-invoicing in Pakistan and other developing countries. 
 
De Boyrie, Nelson and Pak (2007) examined trade mis-invoicing between 2000 and 2005 from 
Africa to the USA. The findings indicate a 60 percent rise in defaults, as low export rates encourage 
tax evasion and money laundering. Higher priced imports make capital flows and can be used for 
47 
 
University of Ghana http://ugspace.ug.edu.gh 
hiding illegal fees.  To boost the current account balance, a nation that faces constant current 
account deficits will take on a devaluation. Alternatively, it can internally raise capital by 
facilitating a private sector switch. This can be achieved through direct appropriation/ 
nationalisation of private assets or inflationary tax-generating seignorage revenue. In either case, 
the private sector will be encouraged to move its assets outside government control. 
 
Cones (1987) and Williamson (1987) argue that higher GDP growth rates signal the presence of 
attractive investment opportunities at home which is an incentive for firms to undertake more 
domestic investment. This reduces the incentive to mis-invoice and send capital abroad.  
 
2.6 ABUSIVE TRANSFER PRICING/TRANSFER MISPRICING 
Transfer mispricing is a form of mis-invoicing by similar parties in different countries, such as 
different multinational corporate industries (Wang, 2007). In this way, companies can pass income 
to low-tax countries from their home country and avoid higher tariffs. The transfer price concerns 
the pricing of transactions between associated companies (OECD, 2010). The related business 
refers to an individual's parent, partner, trustee of a trust, a partnership in which the person is a 
partner and that controls at least 50% of the rights to the income or capital of the association and 
a company managed on its own or by other partners. Transfer prices are a method of price 
determination for products or services exchanged between parts of companies' multinational 
groups. 
 
Transfer pricing is often used Multi National when interests travel through numerous jurisdictions 
in Multi National Company (MNC) functions. It is normally usually seen as a reaction to external 
48 
 
University of Ghana http://ugspace.ug.edu.gh 
pressures, such as government regulations (Eden, 2012). The 'over-or under-invoicing' of such 
transfer prices is an abuse of such a mechanism. Manipulating the transfer price is an illegal flow 
that jeopardises developing states' capacity to establish capable and legitimate institutions. The 
transfer price must be understandable to determine Transfer Price Manipulation (TPM). 
 
Hirschleifer (1956) provides one of the first extensive transfer pricing assessments. It argues from 
a rigorous neo-liberal perspective that transfers between international affiliates must react 
rationally and efficiently to market conditions and that transfer prices must comply with the rules 
of fully comparative markets. Whilst he acknowledges that there are no completely open markets, 
he prescribes transfer prices under more rational assumptions. In the words of Hirschleifer (1956), 
only if the product being transferred is produced in a comparative market, i.e. comparative in the 
theoretical sense, is the correct market price transference only if the single producer considers 
itself as being big enough to control demand through its own decisions concerning production. 
Suppose the need is unrivalled or there exists no need for the transmitted product. In that case, the 
appropriate practice is to sell it at a marginal cost or at any price between the marginal and marked 
costs, in the most general case, at some simplification conditions. 
 
In general, it makes economic sense to conduct subsidiaries on sites close to fixed capital 
operations and in the proximity of resources themselves, to ensure sufficient marginal pricing 
environments (depending on the requirements of such items as untrained or skilled labour, for 
example). Hirschleifer (1956) concludes that apart from either the demand or the marginal price 
of transfer rates (the closest pricing) will distort the order in a manner that can yield suboptimal 
results for the various trading subsidiaries. 
49 
 
University of Ghana http://ugspace.ug.edu.gh 
The most commonly applied anti-profit shifting laws are transfer price laws and exist in 10 
LDCs/LICs, 26 LMICs and 31 UMICs (Ernst & Young, 2010). They determine intra-firm prices 
to align with the arm's length principle. The OECD Pricing Directives control most transfer pricing 
laws in developed countries. Whether rules for a community of high-income countries are always 
in line with the interests of developing countries is unclear. Brazil is an example of a country that 
does not comply with the OECD Guidelines on the transfer pricing rules to suit its scheme better. 
Generally, it is challenging to enforce transfer price rules because there are often no comparable 
internal or external cases where correct transfer pricing can be established. The tax authorities have 
no technical expertise and capacity to effectively audit corporate taxpayers' transfer pricing 
documentation reports. Most multinational companies (MNCs) competing in Africa aim to reduce 
their tax burdens, make enormous profits, and transfer revenues to their country of origin. 
 
Through tax evasion, corruption and inadequate transfer prices, and maladministration, the 
continent lose billions of dollars annually, leading to Africa's illegal financial flows. Previous 
studies have shown that if adequate flow control mechanisms are in place, African revenues could 
increase between 60 and 65%. Recent reports have also demonstrated that Africa loses around 
USD58 billion annually on exports and imports-related activities (UNECA 2015). Some 
multinational corporations (MNCs) usually benefit from concessional tax through corrupt African 
government activities. Most MNCs tend to maximise profits while illegal financial transactions 
destroy Africa's economy. These corrupt actions are why Africa remains very poor and incapable 
of competing with the world's best countries. 
 
 
 
50 
 
University of Ghana http://ugspace.ug.edu.gh 
2.7 OVERVIEW OF ANTI MONEY LAUNDERING 
Money laundering (ML) is the practice of making a substantial sum of unlawfully obtained money 
(from drug trafficking, terrorist activities, or other major crimes) appear to have come from a 
legitimate source. The criminal act of concealing money from illicit operations and converting it 
into legitimate money (Le-Khac et al. 2016; Syed et al. 2019) changes the clandestine character of 
money. Money laundering is the process through which illicit money is transformed into white and 
clean money (Cotoc et al. 2022). If done correctly, it helps criminals to keep control of their gains 
and, eventually, to offer a respectable cover for their source of income. The International Monetary 
Fund (IMF) and World Bank are now involved in money laundering issues as a result. 
 
Money laundering is critical in assisting the objectives of drug traffickers, terrorists, organised 
criminals, insider dealers, tax evaders and many others who seek to escape the type of attention 
from authorities that sudden income from unlawful operations brings. These criminal businesses 
strive to gain money and influence through illicit activity before infiltrating normal society and 
altering the parameters of the bargain. They create millions of dollars for the enterprise's members 
and allow its associates to live extravagant lifestyles built from the pain and sorrow that their illicit 
activity causes (Kumar, 2012). 
 
According to the literature, the money laundering process consists of three stages: placement, 
layering and integration (Demetrias and Vassileva 2020). The initial step of money laundering is 
the most dangerous since it includes the entrance of illegal gains into the financial system, with 
the chance of being identified by authorities (The World Bank and Schoot , 2003; Jaara and 
Kadomi 2017; Jayantilal et al. 2017). Stratification denotes laundering and obscuring the unlawful 
51 
 
University of Ghana http://ugspace.ug.edu.gh 
source through many transactions (Demetrias and Vassileva, 2020), and in the last step, 
integration, money or income are reintroduced into the legal economy (Teichmann et al. 2020). 
Money laundering can be done locally ("adding cash to the cash registers of a cash-intensive firm") 
or globally (depositing filthy money by criminals in the bank and targeting the financial system to 
lose the unlawful mark with the aid of offshore corporations (Ferwerda et al. 2020).  
 
The Reagan US Presidential Commission on Organised Crime (1986) spearheaded efforts to 
encourage investigators to "follow the money," but academic criminologists and economists had 
little to no influence on the anti-money laundering movement, despite any small influence some 
individuals may have had on the fight against organised crime.  
 
Despite the private reservations of some law enforcement officials and professionals, academic 
criticism and/or empirical research had no discernible impact on slowing the anti-money 
laundering movement down. Even within the EU, however, there are differences in emphasis and 
structure related to national cultures and legal principles (Vogel and Maillart 2020). Since the 
process of identifying and recovering proceeds of crime can frequently last longer than a pre-
determined or elapsed time year, the data reviewed would require a longitudinal flow method of 
accounting rather than an annualised method. This fact has, however, received relatively little 
attention to date. 
 
The Financial Action Task Force (FATF), an intergovernmental organisation founded in 1989 on 
the initiative of the G7 to develop policies to combat money laundering, makes no mention of 
doing a formal risk assessment. The standard merely requires such nations to 'identify, analyse and 
52 
 
University of Ghana http://ugspace.ug.edu.gh 
comprehend' hazards, not to generate a written record. Thus, a country may have a range of papers 
or instruments that, when considered together, may fulfill the requirements of the FATF guidelines 
but do not constitute a genuine National Risk Assessment (NRA). Many Global South countries 
in Asia and elsewhere do not publish their NRAs, and those that do so only publish sanitised 
versions: it is debatable whether the classified versions are significantly more informative, not 
least because, while they may contain more detailed cases and names, it is unclear what data they 
could collect.  
 
There is no proof that any of the global statistics produced and repeated endlessly (e.g., UNODC 
2011) as "facts by repetition" are used operationally as a baseline for evaluating the effectiveness 
of AML or of any other policy; if they were, as with the various "wars on drugs" have been, the 
policy framework may have to be abandoned or significantly altered. However, performance 
statistics are still required even in those nations where evidence-based policing is not the norm, 
even if many of these nations struggle to effectively use analytical data as proxies for efficacy. 
There are several articles that address the issue of suspicious money laundering transactions from 
a theoretical perspective, as well as articles that address the same issue through case studies 
(Naheem 2016; Raza et al. 2020; Gilmour 2020), but an approach from a statistical perspective of 
the suspicious transactions available in annual reports for recent years has not been found. The 
findings of this study assist governmental and non-governmental organisations by emphasising 
areas that require immediate action to mitigate the consequences of ML. Setting global guidelines 
or having nations apply them and demonstrate compliance with the requirements is insufficient. 
They must also have an impact and improve the efficiency of AML and CFT, the execution of 
relevant EU legislation in line with international standards in the fight against money laundering, 
53 
 
University of Ghana http://ugspace.ug.edu.gh 
as well as how it is implemented, must be examined on a regular basis. We could not find an 
adequate statistical analysis of suspicious transactions in our inquiry. 
 
The underlining importance of international money laundering legislation in the battle against 
money laundering was studied, based on data collected and analysed from 20 reports of European 
Union Member States' FIUs. This study shows an increase in suspicious transaction reports (STRs) 
received by anti-money laundering national bodies between 2018 and 2019, as well as the newest 
money laundering scheme of national Financial Intelligence Units (FIU) because of European 
Union measures and their transposition into national laws. 
 
The time frame chosen for analysis coincides with the final term for the implementation of the 
latest European AML Directive, the directive issued in June 2018, with a deadline for transposition 
into national legislation of January 2020; thus, the statistics available in the field for 2018 and 
2019 more or less reflect the result of the new European measures. The AML/CFT frameworks 
operate in tandem to maintain the financial system's safety, integrity and soundness, as well as 
depositor protection. To run an effective AML/CFT framework, institutions should tighten their 
regulations, which may contain duties that fit the capacity of both public and private institutions, 
and law enforcement should be placed in both institutions to preserve order. 
 
In practice, the state should adopt some flexibility in the application of AML/CFT measures, which 
would increase the chances of attracting more segments of the population into the formal financial 
system, making it easier to monitor and detect any movement of funds or attempt to conceal the 
proceeds of crime. The proper AML/CFT procedures, according to Shehu (2012), to minimise the 
54 
 
University of Ghana http://ugspace.ug.edu.gh 
impact of any financial crisis on the financial system. Lack of financial inclusion, particularly 
restricted access to financial services, usage of informal channels, and the predominance of major 
informal service providers, among other factors, make tracing and monitoring transactions harder, 
resulting in a weaker AML/CFT regime. 
 
AML/CFT procedures are mechanisms put in place to support financial institutions' effective 
intermediation function, defend the integrity and soundness of the financial system, and guarantee 
that only real economic operations are done to promote economic growth and development. 
 
Money laundering is a worldwide concern that demands policymakers' immediate response to 
protect financial institutions from criminals. A jurisdiction is necessary to build a system to 
regulate or limit the impact of multifaceted crimes on the global economy. Global crime prevention 
poses a challenge not just to jurisdictions but also to global norm setters. The banking sector and 
other regulated entities bore the brunt of the consequences of these misdeeds. Anti-money 
laundering/counter-terrorism financing (AML/CFT) supervision is a developing field; hence, a 
solid AML/CFT supervisory regime is required to support worldwide endeavors in this area. The 
implementation of an effective risk-based AML/CFT supervision of financial institutions as well 
as specified non-financial businesses and professions is important for AML/CFT regulators to 
strengthen the AML/CFT regime of a country. 
 
2.8 CHAPTER SUMMARY 
This chapter reviewed relevant literature on trade mis-invoicing, abusive transfer pricing and the 
determinants of trade mis-invoicing. This chapter explained various relevant concepts such as trade 
55 
 
University of Ghana http://ugspace.ug.edu.gh 
mis-invoicing, abusive transfer pricing and trade discrepancy. Also, the multiple ways of 
measuring trade mis-invoicing such as the PCTG method and the Price Filter method, were 
discussed with their limitations to provide a thorough understanding of the study. The chapter also 
reviewed the factors which affect trade mis-invoicing and concluded with an overview of money 
laundering.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56 
 
University of Ghana http://ugspace.ug.edu.gh 
CHAPTER THREE 
 
METHODOLOGY  
3.1 INTRODUCTION 
This chapter presents the methodology used for the study. It begins with the qualitative research 
methodology used to analyse the value chains for the commodities and identify the risks for IFFs. 
Then it outlines the methods used for estimating illicit financial flows and ends with the 
methodology for measuring the determinants of Illicit Financial Flows. 
 
3.2 VALUE CHAIN ANALYSES AND IDENTIFICATION OF RISKS FOR IFFS IN 
SELECTED COMMODITIES 
This thesis adopted mixed research methods, i.e., expert interviews and literature searches, to 
understand and identify the risks of trade mis-invoicing in the value chains of the selected 
commodities. In addition, the methods adopted allow the study to triangulate relevant data sources 
and existing analyses and set up the assumptions underpinning the estimations of trade mis-
invoicing in the commodities.  The combination of quantitative and qualitative approaches offers 
more significant insights into research concerns and problems than the individual use of one tool 
(Creswell, 2012). Similarly, Salmona, Kaczynski, and Smith (2014) posit that the service of 
qualitative research allows the researcher to ask different questions and explore multiple forms of 
data in unique ways.  
 
The purpose of the qualitative research was to gain knowledge of and evaluate expert perspectives, 
existing research, new data and economic incentives for trade mis-invoicing in the commodity 
57 
 
University of Ghana http://ugspace.ug.edu.gh 
export sector in Ghana. Data was gathered through desktop research and semi-structured 
interviews with experts at related private and public sector institutions. Purposive sampling, 
defined by Etikan, Musa and Alkassim (2016) as a sampling technique suitable for selecting 
interviewees based on their expertise in a particular field, was used. 
 
The institutions and interviewees were therefore selected based on their special positions in the 
gold, bauxite, manganese, and cocoa sectors in Ghana. They included private sector trade 
associations and government agencies responsible for managing natural resources, foreign trade 
governance, and tax and revenue authorities. Expert interviews were suitable for the research due 
to the technical knowledge needed to support the study of the risks and magnitudes of trade mis-
invoicing. A total of twenty-five people were interviewed. The interviewees were primarily at the 
level of senior management and as such could be considered experts in their respective fields. 
 
Wide-ranging additional desktop research was used to identify current research studies and data to 
inform the interviews and statistical analyses. Data for the desktop analysis was collected from the 
websites of the listed institutions and their annual reports, which were accessible online. Journal 
articles and current literature on keywords, such as product supply chain analysis, trade mispricing 
and foreign trade, were also used. 
 
For the gold, manganese and bauxite sectors, the organisations contacted were the Minerals 
Commission (the regulators), Ghana Chamber of Mines (responsible for large-scale mines) and 
Precious Minerals Marketing Company (PMMC) (the national assayer). Additionally, Ghana 
Revenue Authority (Customs Division), Ghana Extractive Industries Transparency Initiative 
58 
 
University of Ghana http://ugspace.ug.edu.gh 
(GHEITI) and ISODEC were included to enrich the source of data. With respect to the cocoa 
sector, COCOBOD, Cocoa Marketing Company (CMC) and Ghana Export Promotion Authority 
(GEPA) were contacted. 
 
The following institutions, with overall responsibility for the governance of the natural resource 
sector in Ghana: Bank of Ghana, Ghana Revenue Authority (Customs Division) and Ghana 
Investment Promotion Centre (GIPC), were contacted.  
 
There was a positive response of about 95% from all the organisations contacted. However, there 
were no interviews with representatives of the individual gold, bauxite, and manganese mining 
firms, the Ghana Free Zones Authority, and individual cocoa paste processing firms.  
 
 
 
 
 
 
 
 
 
 
59 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 3:1 displays the list of institutions contacted for information and validation purposes. 
Table 3.1: List of Institutions Contacted 
Cocoa Beans and Cocoa Paste 
Role of Institution Name of Institution 
Production Ghana Cocoa Board (COCOBOD) 
Transportation 
Export Cocoa Marketing Company (CMC) (COCOBOARD), 
Ghana Export Promotion Authority (GEPA) 
Gold, Bauxite and Manganese 
Role of Institution Name of Institution 
Regulator Minerals Commission 
Private Sector Organisation Ghana Chamber of Mines, Freight Forwarding Agency, 
Gold Exporting Agencies 
Assaying/Valuation Precious Minerals Marketing Company (PMMC) 
Civil Society Organisations Ghana Extractive Industries Transparency Initiative 
(GHEITI), Integrated Social Development Centre 
(ISODEC) 
Natural Resource Sector Governance 
Purpose of Institution Name of Institution 
Regulator/Tax Authority Customs Division (Ghana Revenue Authority) 
Regulator/Central Bank Bank of Ghana 
Business Promotion Ghana Investment Promotion Centre (GIPC) 
Data Source: Expert Interviews 2018, 2019 
 
60 
 
University of Ghana http://ugspace.ug.edu.gh 
3.3 MEASURING ILLICIT FINANCIAL FLOWS 
IFFs, by their very nature, are difficult to measure (Collin, 2020). As a result, the trade gap analysis 
method has become one of the most widely used methods for measuring IFFs. This works on the 
principle of estimating the trade gaps. However, there have been a lot of criticisms about the use 
of the trade gap methods for measuring IFFs because trade gaps/discrepancies could exist for other 
reasons aside IFFs (WCO 2018).  
 
Trade Gap Analysis is a method of measuring IFFs, which was first developed by Bhagwati 
(1964a).  The concept is very simple: the declared price and quantity of exports from one country 
will, after accounting for shipping and insurance costs, be compatible with the declared price and 
quantities when that good arrives and is registered as an import. Shipping and insurance costs 
(which are generally non-registered as aggregated trade statistics) or a mistake in reporting 
import/export value or quantity are the only valid distinction between the two records. 
 
As described by Bhagwati (1964b), the value (or quantity) of an export or import may be declared 
under or over an incentive. For example, if an agent wants to move money secretly from Malawi, 
he might charge USD100,000 for importing cell phone shipments from the USA to an intermediate 
outside Malawi. There are various incentives ascribed to over-/under-reporting exports/imports, 
which include: 
(1) over-invoicing imports—retaining money abroad, avoiding capital controls (by obtaining 
excess foreign exchange), reducing taxable profit; 
(2) under-invoicing imports—evading customs duties; 
(3) over-reporting exports—taking advantage of export credits; 
61 
 
University of Ghana http://ugspace.ug.edu.gh 
(4) under-invoicing exports—avoiding export tariffs, retaining money abroad, avoiding 
capital controls (by obtaining excess foreign exchange). (Collin 2020) 
The "gross exclusion reversals" method of Global Financial Integrity (GFI) explores mirrored 
statistics (i.e., which country A says it exported from country B and which country B declares it 
imported from country A, respectively) between developed countries on the basis that the values 
reported in the first case to customs officers are most likely valid. The GFI method tests the "gross 
exclusion reversals"(GER) process. GER takes into account that, due to shipping costs and 
insurance costs, the two values mentioned shall be automatically divided into 6% of export value 
when real details on those costs are not available. Based on the incentives mentioned above, GER 
assumes that any difference in reflected trade statistics indicates over-valued or under-valued 
exports. The approach presupposes that shipping costs will be corrected. Note that the assumption 
of 6% is invariant in respect of export distance travel, so it is possible if distance leads to increased 
costs of shipping and vice versa, that under-invoicing is reported (a false positive). 
 
GFI uses GER for the first time to add up the total IFFs amounts due to commercial money 
laundering for each developing country and the wealthier counterparts to estimate countries and 
global figures. The IMF uses both aggregated data from the database of the IMF Trade Statistics 
and commodity-specific data from the UN COMTRADE database in its latest report (GFI 2017). 
It then uses the amount to estimate the share of trade in the country involving unlawful transactions 
and increases that share to an aggregate value for unlawful outflows by aggregate estimates. When 
summarizing figures at the country level, the worldwide projected money laundering is focused on 
trade. In 2015 alone, GFI figures put the total volume of money left by developed countries for 
mis-invoicing at approximately 85 per cent of the total amount (598–807 billion dollars per year) 
62 
 
University of Ghana http://ugspace.ug.edu.gh 
(Salomon 2018). In addition to GFI, several research outfits and NGOs have developed similar 
figures for each region, country group, and category of goods (Nicolaou- Manias, 2016).  
 
Those methods were strongly criticised for suggesting more innocent reasons why the 
representative trade statistics could differ. The reasons include price or quantity errors, differences 
in product categorization between countries, differences in rules or accountability on reporting, 
discrepancies between real shipping and insurance costs and the 6% assumption used in the GER 
method (Johannesen and Pirttilä, 2016; Nitsch, 2016). Some of the same critiques have challenged 
the claim of GER that figures on trade are calculated from rich countries without (or at least with) 
error. As a result of these valid critiques, trade disparities are likely to be both misleading and 
unnecessarily inflationary.  
 
Some have also pointed out that re-exporting through bonded warehouses is not sufficiently 
protected by the GER system. For example, GFI corrects its re-export figures between China and 
Hong Kong, but the issue will likely still affect other bilateral estimates (Forstater 2016). Such 
problems are not strictly theoretical, as high-profile cases of the GER cycle which have led to 
substantial overestimation of IFFs have been identified. Most significant in 2016 was the study by 
the United Nations Trade and Development Conference (UNCTAD), which revealed that 67% of 
South Africa's gold exports were mis-invoiced and IFFs amounted to about 80% from 2000 to 
2014.This disparity was discovered later because of a gap in reporting requirements that prevented 
accurate registration of gold exports in the COMTRADE database (Forstater, 2017a and b). 
 
63 
 
University of Ghana http://ugspace.ug.edu.gh 
There are two main methods for estimating trade mis-invoicing – the Partner Country Trade Gap 
(PCTG) Method and the Price Filter Method.  
 
3.4 THE PARTNER COUNTRY TRADE GAP (PCTG) METHOD 
The partner-country trade gap methodology compares partner countries’ import and export reports 
for the same trade flow. Trade gaps differ between the partner’s reported imports and the partner 
countries’ reported exports. Following the literature, a correction factor of 10% is applied to the 
export values to account for the cost of freight and insurance (Baker, 2014; Berger and Nitsch, 
2012).  For each commodity, the gaps were calculated according to the following equations  
Export Trade (ET) gaps in a year t are:  ETij,t = IMij,t − β ∗ EXij,t…………………………Eqn 1 
EXij,t is partner j’s reported export value, IMij,t is country i’s reported import value from partner 𝑗 
and 𝛽 is the freight and insurance factor. The partner j’s imports are compared to the interest 
country i’s corresponding exports for the same trade flow. A positive trade gap value of 𝐸𝑇 
indicates export under-invoicing while a negative value indicates export over-invoicing. 
(Bhagwati, 1964b; Ndikumana, 2016). 
 
This method is implemented as follows:  
First, annual import data (measured using cost plus insurance and freight or c.i.f. method) are 
converted to the free-on-board basis using a standard factor of 10 per cent used by the International 
Monetary Fund’s (IMF) Direction of Trade Statistics (DOTS). Next, the import and export 
discrepancies (I.D. and E.D., respectively, as denoted below) are calculated using the following 
formulae:  
I
ID j,tij,t = − Xj,t……………………………………………………………………………Eqn 2 r
64 
 
University of Ghana http://ugspace.ug.edu.gh 
Ip,t
EDij,t = − Xp,t ………………………………………………………………………….Eqn 3 r
Where:  
Ij,t is the imports by country j from the partner country p at time t  
Ip,t is the Partner country p’s imports from the country j at time t  
r  is the Standard freight and insurance cost adjustment factor of 10 per cent,  
Xp,t is the Partner country p’s exports to country j at time t  
Xj,t is the Country j’s exports to partner country p at time t 
These discrepancies or trade gaps are usually calculated at the aggregate level, including all traded 
product categories, but they can also be calculated at the product level (based on the Harmonised 
Commodity Description and Coding System or HS code) subject to the availability of 
disaggregated data. Most recent applications of this approach include the empirical literature on 
IFFs conducted by advocacy organisations and policy institutions. 
 
3.5 THE PRICE FILTER METHOD 
Following the criticisms of the PCTG method, some researchers developed another method for 
estimating trade mis-invoicing in international trade known as the price filter method. (eg. Pak & 
Zdanowicz, 1994; and Hong et al., 2014). This method makes use of transaction-level micro-data 
to estimate trade mis-invoicing. It also uses the Harmonised Commodity Description and Coding 
System (HS Code) at the 10-digit level.  
 
Additionally, the price filter method estimates illicit financial flows based on export/import 
deviations from some reasonable price range instead of using mirror trade statistics. Zdanowicz 
(2009) examines US exports and imports and argues that transactions with prices above any 
65 
 
University of Ghana http://ugspace.ug.edu.gh 
distribution margin are reasonably the result of unlawful behaviour (e.g. 50% of the retail price, or 
upper / lower quartile). Using this analytical method combined with corporate tax rate data, 
Christian Aid (2009) calculated the loss of USD 160 billion per year of developed countries as a 
result of consumer mispricing. There is a problem in this method in which two different forms of 
transactions are conceptually combined: commercial mis-invoicing (over / under-invoicing at one 
end of the transaction and reporting at the other end) and transfer mispricing (where the price is 
accurately stated at both ends of the transaction but is skewed to impact the position of the 
transaction report). An unexpected low export price may be triggered by an artificially low sale 
price or by a low bill. 
 
As the price variance estimates of the IFFs obtained under this approach can increase in a given 
product category, it is difficult to determine which percentage of such estimates are driven by 
natural price deviations and which estimates are actually driven by some form of illegal flow. As 
with studies of trade disparities, data errors also appear as illegal sources. Forstater (2015) notes 
that price outliers for trading data (e.g. a bucket at USD 973 used as an example of mis-invoicing 
for many non-profit companies) are likely to be explained partially by errors in the registration of 
either the price or the quantities of shipment. Likewise, products that have a higher / lower quality 
than or different from the desired classification could be flagged erroneously as prices that are 
higher / lower than normal. 
 
Given these limitations, it is possible that the study of significant anomalies in exchange rates 
would be a valuable way of identifying forensic behaviours used by actors such as customs 
officers.  
66 
 
University of Ghana http://ugspace.ug.edu.gh 
There are two main methodologies under the price filter analysis. These are the: inter-quartile 
range filter and arm’s length price filter. For this thesis, the inter-quartile price filter will be used. 
This is due to the unavailability of free-market prices for the commodities in question in order to 
use the arm’s length filter methodology.  
 
An endogenously calculated interquartile range (IQR) price filter method is calculated. This 
method is motivated by the United States Internal Revenue Service (IRS) transfer pricing 
regulation (Internal Revenue Code 482) which specifies that an interquartile range is an acceptable 
arm’s length transaction range (Hong et al., 2014). The OECD Transfer Pricing Guidelines also 
mention that the interquartile range or other percentiles can be used to help enhance the reliability 
of any transfer pricing analysis (paragraph 3.57, OECD, 2017). All transactions valued in the top 
and bottom quartile of these distributions are designated to be abnormally valued.  
 
Accordingly, any transaction value which exceeds the 75th percentile or falls below the 25th 
percentile of the observed price distribution is designated to be abnormally priced. The under or 
over-valued amounts for each transaction is then calculated as follows: 
Undervalued amount = Quantity x MAX(0, LoQ - P)………………………………Eqn 4 
Overvalued amount = Quantity x MAX(0, P - UpQ) .………………………………Eqn 5 
where: 
P = Declared price (unit value implied in quantity and value in each trade record) 
LoQ = Lower-quartile price calculated using price distribution over the previous 365 days 
UpQ = Upper-quartile price calculated using price distribution over the previous 365 days 
 
67 
 
University of Ghana http://ugspace.ug.edu.gh 
It is relatively straightforward to observe that since the interquartile price range is endogenously 
estimated using the observed price distribution, this hypothesis will be rejected by design for a 
certain proportion of transactions. Therefore, these estimates of trade mis-invoicing should be 
interpreted carefully and supplemented with a further discussion regarding the product, price, and 
individual market characteristics. For example, the level of purity of precious metal and any 
contemporaneous political, economic or environmental shocks may play a key role in determining 
whether the observed transaction price falls within the interquartile price range during a given 
period. 
 
3.5.1. Arm’s Length Price Range for Gold Exports 
Gold doré produced in Ghana ranges between 67:33 and 100:0 gold – silver split (after accounting 
for a maximum of 5% of impurities like lead, copper, and arsenic). Accordingly, the first step is 
to calculate the relevant benchmark prices for the various proportions of gold and silver found in 
the gold doré bars produced in Ghana. Secondly, the commodity-specific factors which result in 
further variation of observed export values to deviate from the free-market reference prices are 
assessed. This information is combined to estimate the arm’s length price range between PLow 
and PHigh. 
 
Benchmark Price for Gold: In order to establish the benchmark prices for different levels of 
gold-silver mix, the firm names in the gold exports data from Ghana Revenue Authority are 
merged with mine-level production information from the Metal Focus Gold-Silver Dore Service 
database. However, not all the exporters can be matched to individual gold mines due to the 
presence of intermediate buyers and logistical firms in the GRA export data. In these cases, 
conservatively the minimum and maximum gold purity levels observed across all gold mines are 
68 
 
University of Ghana http://ugspace.ug.edu.gh 
used to estimate the benchmark prices. In all cases, the gold and silver content is adjusted to 
account for a maximum of 5% permitted impurities in the doré using the following adjustment: 
(gold/silver purity*0.95)/100. 
Accordingly, the benchmark is calculated as follows: 
Case 1 - GRA exports data matched with mine-level gold-silver 
content: 
 
Benchmark price = (the price of gold*gold content) + (the price of silver * 
silver content)……………………………………………………………………Eqn 6 
 
 
Case 2 - GRA exports data which cannot be matched with mine-level gold-silver content. In this 
case, a minimum and maximum range for benchmark prices are calculated as follows: 
 
Maximum benchmark price = (price of gold * maximum gold content) + (price of silver 
*minimum silver 
content)………..…………………………………………………………………….Eqn 7 
Minimum benchmark price = (price of silver * maximum silver content) + (price of gold * 
minimum gold content)……………………………………………………………………..Eqn 8 
Upper bound of arm’s length price range (PHigh): The arm’s length price range is estimated 
by first calculating the upper bound using the following commodity-specific factors based on 
estimations by interviewed gold sector experts: 
  
69 
 
University of Ghana http://ugspace.ug.edu.gh 
1. Transport, storage, and insurance costs (negligible impact): These costs are included in 
all export transactions, however, their impact on the total value of precious metals like gold is 
negligible i.e. maximum of 2% of the total value. This decreases further for larger volume 
transactions (Personal Communications, A. Tawiah). 
 
2. Market conditions and contract terms (up to 10%): Export prices also fluctuate due to the 
terms of underlying contracts which include different types of transactions (spot and futures 
contracts). This also reflects differences in price bargaining power between trading companies. 
Favorable forward/futures sales can determine gains of the doré exports against spot prices used 
as a benchmark. Also, Ghanaian Cedi – US Dollar exchange rates and prevailing interest rates 
affect export prices of gold from Ghana (Personal communications, C. Nyarko). 
 
Based on the above analysis, an estimate of 10% above the previously calculated benchmark 
prices is used as the upper bound of the arm’s length price range for gold: 
Upper bound of arm’s length price range = (Benchmark 
price*1.10)…………………………………………………………………………………....Eqn 9 
 
 
Lower bound of arm’s length price range (PLow):  The assumption for the lower bound of 
the arm’s length price range as arrived at as follows: 
1. Market conditions and contract terms (up to 10%): Contract types and market conditions 
have a negative impact in terms of reducing the observed export prices below the calculated 
benchmark prices. Therefore, a total downward impact of 10% is assumed to calculate the lower 
bound of the arm’s length price range as follows: 
70 
 
University of Ghana http://ugspace.ug.edu.gh 
Lower bound of arm’s length price range = (Benchmark 
price*0.90)…………………………………………………………Eqn 10 
 
 
Declared export values falling above the upper bound arm’s length range indicate overvaluation 
while those below the lower bound arm’s length range indicate undervaluation of exported gold. 
 
3.5.2. Arm’s Length Price Range for Cocoa Beans 
Unlike gold, the empirical methodology for estimating mis-invoiced cocoa beans exports does 
not require detailed information on different types of purity levels. This is because Ghana exports 
only Grade I and Grade II cocoa beans which are generally considered as high-quality beans, 
even above standards set by the international market, hence, attracting premiums in addition to 
the international market reference prices (Quarmine et. al., 2012; Abbott, 2013).  In order to 
ensure maintenance of this internationally recognised high-quality cocoa beans standards set by 
COCOBOD, Cocoa Marketing Company (CMC), a subsidiary of COCOBOD, have (by law) been 
given the sole right to sell and export cocoa beans from Ghana as well as perform the take-over 
function within the internal marketing system. 
 
The cocoa marketing procedure in Ghana is unique with its partly liberalised system where 
internal marketing undertaken by LBCs are privatised, although with strict oversight by CMC. 
  
Benchmark Price for Cocoa Beans Exports: The London Futures Prices are used as the 
benchmark prices to estimate trade mis-invoicing in cocoa beans. This is because export prices 
of the beans are negotiated based on these prices. The Intercontinental Exchange (ICE), London, 
particularly represents delivery in Northern Europe and serves as reference prices for West 
71 
 
University of Ghana http://ugspace.ug.edu.gh 
African cocoa; hence, CMC’s trading of cocoa beans, especially to foreign buyers, is based on 
these prices. In the agreement outlined, buyers and sellers present the futures contract as a 
reference price and then negotiations are undertaken in relation to differential premium or 
discount depending on quality, default and counterparty risks in delivering the quantities for 
which prices are fixed. 
  
Ghana’s institutional reputation with regard to quality and counterparty risk, for instance, allow 
for sales that are usually conducted six to 12 months (or sometimes 15 to 18 months) prior to 
delivery. The main reason for the futures sales is that the sales serve as collateral for the syndicated 
loans acquired to pay LBCs for their cocoa beans, provide research and extension services to 
farmers as well as funds for the general administration of all cocoa institutions in the country 
(Tröster et al., 2019; Personal Communications, CMC Personnel). Based on assumptions 
guided by the qualitative research conducted, i.e. communications with experts in the industry 
and desktop research, an arm’s length price range is subsequently estimated by focusing on market 
conditions and contract types as well as transport and insurance costs. The upper and lower bounds 
arm’s length price ranges are set as follows: 
Upper bound of arm’s length price range (PHigh):  The following considerations are used to 
estimate PHigh for cocoa beans exports from Ghana: 
 
1. Market Conditions and Contract Terms: With the peculiar cocoa supply chain in Ghana, 
varied sales strategies undertaken by the traders based on instinct with experience, plausible 
portions of estimated abnormal pricing are assumed to be a reflection of premiums earned as a 
result of a combination of an established guaranteed delivery of sold cocoa beans on time, quality 
72 
 
University of Ghana http://ugspace.ug.edu.gh 
of beans, market power and counterparty premium. For example, CMC/COCOBOD selects top-
grade quality and size of cocoa beans for export that attracts roughly 3 – 5% price premium on the 
world markets. Certifications such as Organic, Fairtrade, UTZ, and traceability also attract a price 
premium of 150 USD per tonne which is about 5.4% of the average value of London’s Futures 
Prices for the period 2011 – 2017 (Gilbert, 2009 in Kolavalli and Vigneri, 2011; Dand, 2011). 
 
Some reasons given for the premiums include slightly higher-than-average fat content, low levels 
of debris that result in higher cocoa butter yields, and low levels of bean defects that generate 
cocoa liquor flavor preferred by some end users (Kolavalli and Vigneri, 2011). Premiums have 
been known to go as high as 16.5% of the market reference (Personal Communications, 
CMC/COCOBOD Personnel; Stakeholder Meeting at COCOBOD). 
 
2. Transport, Storage, and Insurance Costs: According to CMC, cocoa beans are sometimes 
exported with pre-financed shipments by COCOBOD upon customers’ request. On these 
occasions, the beans are exported under Cost, Insurance and Freight (CIF) terms, which usually 
cost around USD 100 per tonne or 10% of export value (Dand, 2011; Personal Communications, 
CMC/COCOBOD Personnel). Also, using the world market price in 1999, Pedersen (2001)  
estimated the transport cost of cocoa exported from  Tema to Rotterdam to be around 13% of the 
market price. 
 
Overall, a 30% increase above the benchmark price is used (i.e the futures market reference prices) 
to set the upper bound for the arm’s length range beyond which export values are considered 
overvalued (trade mis-invoicing) 
73 
 
University of Ghana http://ugspace.ug.edu.gh 
Lower bound of arm’s length price range (PLow): 
Market conditions and contract terms: It is deduced that based on market conditions and 
contract terms that fix the majority of export values ahead of the season in forwarding sales, risks 
such as exchange rates, lower than expected estimated volumes due to crop failures and 
smuggling can lead to sales that fall short of the expected prices due to defaults (Tröster et al., 
2019). Also, CMC exports about 20 – 30% of cocoa beans at spot prices (mostly light crops of 
relatively small sizes (Quarmine et al., 2012)) that are subject to the same risks, especially 
smuggling and speculations.  Discounts due to non-delivery or default of the expected quality 
(which hardly occurs) and size can result in export values below the benchmark market reference 
prices. Thus an approximation 20% deviations from the benchmark is used to constitute the lower 
bound arm’s length price range. Consequently, cocoa beans export values falling below this lower 
bound are considered undervalued. 
 
3.5.3 Interquartile Range Price Filter Analysis 
The inter-quartile range price filter method assumes that values between the 25th and 75th 
percentile of the observed distribution of unit prices for a specific commodity denote the arm’s 
length price range. Any transaction that falls above or below this price range is categorised as 
abnormally valued. Traditionally, this method relies on the inter-quartile range being calculated 
for each calendar year. However, in a methodological innovation from previous studies (for 
example, Hong and Pak, 2017), a dynamic version of the previous method is implemented by 
updating the calculation of the inter-quartile range on a daily basis using the price distribution 
observed over the previous 365 days i.e. a 365-day rolling window estimate of the interquartile 
range. The main advantage is to make the definition of the arm’s length price range more 
responsive to pricing dynamics observed over the course of the year. For example, in the case of 
74 
 
University of Ghana http://ugspace.ug.edu.gh 
agricultural commodities, the observed trade prices may be affected by planting seasons, climatic 
variation and market conditions which do not directly correspond to calendar years used by 
previous studies to calculate the arm’s length price range. 
 
Accordingly, any transaction value which exceeds the 75th percentile or fall below the 25th 
percentile of the observed price distribution is designated to be mis-invoiced. The under or 
overvalued amounts for each transaction is then calculated as indicated in equations 4 and 5 and 
repeated below 
Undervalued amount = Quantity x MAX(0, LoQ - P)……………Eqn 11 
Overvalued amount = Quantity x MAX(0, P – UpQ ………………Eqn 12 
where: 
P = Declared price (unit value implied in quantity and value in each trade record) 
LoQ = Lower-quartile price calculated using price distribution over the previous 365 days 
UpQ = Upper-quartile price calculated using price distribution over the previous 365 days 
 
It is easy to observe that since the interquartile price range is endogenously estimated using the 
observed price distribution, this hypothesis will be rejected by design for a certain proportion of 
transactions. Therefore, the estimates of trade mis-invoicing should be interpreted carefully and 
supplemented with further discussion regarding product, price and individual market 
characteristics. For example, contemporaneous political, economic or environmental shocks may 
75 
 
University of Ghana http://ugspace.ug.edu.gh 
play a key role in determining whether the observed transaction price falls within the 
interquartile price range during a given period. 
 
3.5.3.1. Arm’s Length Price Range for Cocoa Paste, Bauxite and Manganese 
Due to the lack of commonly acknowledged market reference prices for cocoa paste, bauxite and 
manganese exports from Ghana, the interquartile range method is used to calculate its abnormal 
estimates. Furthermore, It was observed that a high degree of export price rigidity in the cocoa 
paste exports data, which suggests advance pricing agreements at constant prices between trading 
partners which do not respond to observed prices from relevant commodities exchanges. 
Therefore, apply the rolling-interquartile range price filter method as described above to 
approximate the arm’s length price range for cocoa paste exports was applied. 
Upper bound of arm’s length price range (P th High): The 75 percentile is used to set the 
PHigh  for cocoa paste, bauxite and manganese exports from Ghana. Thus, export values that 
are found above the 75th percentile are estimated as the overvalued amounts of cocoa paste, 
bauxite and manganese. 
Lower bound of arm’s length price range (PLow): Similarly, the 25
th percentile is set as the 
PLow for cocoa paste, bauxite and manganese exports from Ghana. Calculated values below this 
boundary are declared as undervalued amount of cocoa paste, bauxite and manganese. 
 
3.6 DATA SOURCES 
The data was sourced from the UN COMTRADE database and the Customs Division of the Ghana 
Revenue Authority. Additional data from the Metals Focus Database was used in the case of gold.  
76 
 
University of Ghana http://ugspace.ug.edu.gh 
3.6.1 UN COMTRADE Database 
Data from the UN International Trade Statistics Database was used for the partner-country trade 
analyses. The database contains annual trade transactions between partner countries. The UN 
International Trade Statistics Database (UN COMTRADE) contains commodity and partner-
specific trade statistics provided by the national statistical authorities of each reporting country. It 
contains annual trade data from 1962 onwards. This data is transformed into a standard format 
with consistent coding and valuation using the UN’s processing system. Commodities are 
generally reported according to the latest classification system and then converted to the earliest 
classification. Therefore, if data is received in HS 2012 version, it is converted all the way to the 
earliest classifications such as the HS 1988. COMTRADE recommends recording imports as 
‘Cost, Insurance and Freight’ (CIF) and exports as ‘Free on Board’ (FOB) and most participating 
countries adhere to this recommendation in their data submission process. 
 
UN COMTRADE explicitly states the limitations of the database. Due to confidentiality, 
countries’ trade reports might only include data at an aggregate commodity code. The dataset also 
contains missing values as countries do not necessarily report their data for every year. 
Furthermore, data cannot be perfectly converted into earlier classifications due to, among other 
factors, the addition of new commodities over time. Similarly, commodity classifications can also 
differ between countries and over time. This makes comparability difficult over the period 
considered. Lastly, COMTRADE does not report trade flows below USD 1000, necessarily 
censoring our estimates upwards.  
 
77 
 
University of Ghana http://ugspace.ug.edu.gh 
Finally, the absence of a standardised procedure in recording import origins and export 
destinations weakens the comparability of trade statistics. The term ‘partner country’ does not 
necessitate any direct trading relationship since exports and imports are recorded by either country 
of first consignment or the country of the last consignment, and ‘partner country’ could refer to 
either.  
 
3.6.2 Data from Ghana Revenue Authority-Customs Division 
Ghana Revenue Authority (GRA), Customs Division, provided the transaction-level export data 
for this study. The recorded transactions cover trading activities from 2005 to 2017.  The data also 
contains the weight of each exported commodity, a detailed description of the commodity type, 
and the receiving country of the exported commodity.  The weight is described as net mass, the 
weight of the commodity exported without packaging in the system unit of the HS classification. 
The assigned unit for weight in the data is kilogram (kg).  
 
3.6.3 The Harmonized Commodity Description and Coding System (HS Code) 
The Harmonised Commodity Description and Coding System, commonly known as the 
Harmonised System (HS) or HS code is an international multipurpose product nomenclature 
developed by the World Customs Organisation (WCO). The group consists of approximately 5,000 
product classes, each of them classified by six-digit codes, organised in a legal and logical 
structure, and backed by clearly defined rules for uniform classification. The system is used for 
their customs tariffs and for the collection of international trade statistics by more than 200 
countries and economies. More than 98% of goods are classified as HS in international trade.  
 
78 
 
University of Ghana http://ugspace.ug.edu.gh 
Internationally, products are classified up to the 6-digit level while countries may classify products 
up to the 10-digit level. The first two digits show the chapter of the HS code the specific 
commodity falls under. The third and fourth digits relate to the headings under the specified chapter 
and provide further details about the class of commodity. In addition, the fifth and sixth digits 
denote the subdivisions of the heading. Contracting parties to the WCO HS Code are permitted to 
create extra subheadings to cater for their domestic tariff arrangements. Thus these would differ 
from country to country as an export from Ghana for instance would be considered an import in 
other countries. Individual countries use additional ratings and the tariff ratings can therefore be 
8-10 digits, depending on how accurately the product has been classified. For example, with the 
coming into force of the ECOWAS CET, the seventh and eighth digits have become regional sub-
divisions in West Africa. There is also a different opinion that the last two digits consist of a 
statistical code that helps a country to compile its statistics on the products of trade. 
 
The HS code is key in research on abnormal pricing in illicit financial flows. In the Partner Country 
Trade Gap Method, the 6-digit level of the HS code is used for the mirror trade statistics of the 
commodities the countries in question export and import. Similarly, for the Price Filter Methods, 
the 10-digit level of the HS Code is used for the analysis of the commodities. 
 
One challenge realised especially at the developing country level, is the issue of misclassification 
of commodities. As mentioned earlier, this could potentially cause a lot of problems since it is 
expected that commodities will be classified pretty much the same across all countries. 
 
 
79 
 
University of Ghana http://ugspace.ug.edu.gh 
3.6.4 Free-Market Commodity Price Data: 2011–2017 
 
Daily market price data is from Thomson Reuters Datastream, a database of global financial 
markets and economic indicators. The commodity exchanges’ data used as free-market reference 
prices for the analyses are the London Bullion Market Association (LBMA) for Gold Bullion 
LBM (US dollars per troy ounce) with 99.5% to 99.9% purity levels and London International 
Financial Futures and Options Exchange (LIFFE) prices for the assessment of the raw cocoa 
beans (US Dollars per metric tonne). 
 
3.6.5. Metal Focus Gold Doré Flows Service Database: 2019 
This database provides mine-level information on historical, current and forecasted doré 
production (up to 2030) by company and country. It also contains information on current refining 
location, historic production costs, current mineral reserves and resources as well as the gold – 
silver split of doré production. This information covers 652 mining companies in 77 countries. 
Specifically, the database has information on 16 mines in Ghana owned by Newmont Goldcorp 
Corporation, Golden Star Resources, Kinross Gold, Gold Fields, Perseus Mining, AngloGold 
Ashanti, BCM International, Asanko Gold/Gold Fields and Golden Star Resources. Except for 
Newmont Goldcorp Corporation and AngloGold Ashanti, all the mines listed are co-owned by 
the Government of Ghana (GoG). 
 
This study primarily used the gold – silver split information from this database for companies 
that exported gold doré within the study period.  The gold – silver split recorded are 100 – 0%, 
97 – 3%, 96 – 4%, 86 – 14%, 81 – 19% and 67 – 33%.  Since the data is merged on the company 
level with the GRA dataset where companies have differing gold – silver splits from different 
80 
 
University of Ghana http://ugspace.ug.edu.gh 
mines, the lowest gold – silver split is used for the analysis of that company. Of the gold – silver 
split recorded, up to 5% constitutes impurities such as copper, lead and bismuth, which is 
permissible by refineries; and mines are paid based on the percentages of both the gold and silver 
in a doré (Metal Focus Report, 2019). 
Table 3.2 Main Gold Mines in Ghana, Ownership and Gold-Silver Content in Production 
 
Mine Mine Ownership Gold:Silver 
 
Name Split (%)  
Ahafo Newmont Goldcorp Corporation 100:0  
 
 
Akyem Newmont Goldcorp Corporation 86:14 
 
 
Bogoso Golden      Star      Resources      / 100:0  
Prestea Government of Ghana  
 
Chirano Kinross    Gold    Corporation    / 67:33 
 
Government of Ghana  
Damang Gold   Fields   /   Government   of 97:3  
 
Ghana 
 
Edikan Perseus Mining / Government of 100:0 
 
Ghana  
Iduapriem AngloGold Ashanti
 
 100:0 
 
 
Obotan Asanko  Gold  /  Gold  Fields  / 81:19 
 
Government of Ghana  
Tarkwa Gold   Fields   /   Government   of 96:4  
 
Ghana 
 
Wassa Golden      Star      Resources      / 100:0  
Government of Ghana  
 
Source: Metal Focus Gold Doré Flows Services Database, 2019 
81 
 
University of Ghana http://ugspace.ug.edu.gh 
3.7 DETERMINANTS OF TRADE MIS-INVOICING 
Given the nature of IFFs, they are usually measured by generating an estimate or by taking a proxy. 
This measure is then regressed on a series of observable characteristics that are considered to be 
linked in one way or the other to illicit flows. When a certain characteristic has a major impact on 
the measure of IFFs, and that effect is backed by theory, there are two ways to interpret the result 
— depending on a priori assumptions. The first is to interpret the result as suggesting a useful 
knowledge on IFFs' figures, when the theoretical relationship between the feature in question and 
the latent value of the IFF is assumed to be valid. The second interpretation is that if one’s prior 
was that the measure of IFFs (e.g., the trade-gap measure) picks up meaningful variation in IFFs 
(e.g., trade mis-invoicing), then correlation with a given characteristic (e.g., customs 
administration quality) provides empirical validation as to which characteristics predict IFFs. For 
example, if one’s measure of IFFs is found to increase after an unexpected windfall in natural 
resource rents, but only in countries with weak institutions, then this might contain information on 
where IFFs are likely to be occurring. 
 
Disconcertingly, these two interpretations are somewhat competing: they require priors either 
about the theoretical relationship between the measure of IFFs and the characteristics being 
regressed on or about the validity of the IFF measure in the first place. Furthermore, both 
interpretations rely on an important identification assumption: that the various characteristics being 
investigated are uncorrelated with measurement error in the estimate of IFFs. For example, 
countries lacking institutional capacity might struggle both to detect illicit behaviour and to 
produce error-proof statistics that can be used in IFF estimates (Collin 2020). 
 
82 
 
University of Ghana http://ugspace.ug.edu.gh 
It is therefore important to understand and investigate the determinants of abnormal pricing. This 
is because the resources lost through the abnormal pricing could have been used to make a major 
contribution to economic development. Abnormal pricing leads to “lost resources” as the resources 
leave the country without domestic authorities’ control, and the funds needed for growth and 
development would decrease. (Yalta & Demir 2010). Abnormal pricing also has unfavorable 
impacts on high taxes on the income distribution of society through movement of funds abroad by 
other people. That ultimately undermines the national tax base and reduces public income 
(Farzanegan, 2009).  
 
As afore mentioned, there are several drivers which can be grouped under headings such as 
political, legal and regulatory, economic and financial. Economic and financial drivers include 
corporate income tax rates, tax rate differences, indebtedness, import-export taxes and duties, 
capital controls, interest rates and differentials in interest rates, domestic inflation, real effective 
exchange rate overvaluation (REER), capital account openness, exchange rate regimes, 
macroeconomic uncertainty, high budget deficits, low growth rates, splitting interest rates between 
international and domestic and actual exchange rates. Some political drivers of mispricing include 
patronage, political freedom, accountability, political instability, corruption and public-sector 
reforms while some legal and regulatory drivers are tax disclosure requirements, transparency 
initiatives and industry-specific regulations. 
 
One important political factor that drives mispricing is corruption. Berger and Nitsch (2012) 
studied the connection of the top five world importers from 2002 to 2006 (United States, Germany, 
83 
 
University of Ghana http://ugspace.ug.edu.gh 
China, UK, and Japan) between trade mis-invoicing and corruption. They conclude that trade mis-
invoicing by the partner countries increases with the level of corruption.  
 
Another variable the study focuses on is transfer mispricing. Transfer mispricing refers to mis-
invoicing by related parties in various countries, such as countries and branches of a multinational 
company. This practice allows businesses to move income from the home country to low-taxation 
countries and escape higher taxes. By its very nature, transfer mispricing is difficult to measure. 
This study makes use of some proxy variables to represent transfer mispricing. These are Transfer 
Pricing Risk and Transfer Pricing Rank calculated for various countries.  The Tax Rate Differential 
is another factor that could potentially influence abnormal pricing between the exporting country 
and the partner country.  
 
Another variable this study focuses on is inflation. Kwaramba et al. (2016) found that both the 
higher capital outflow and increase in capital flight are correlated with high inflation. The Real 
Effective Exchange Rate (REER) is also another variable of interest.  Kwaramba et al. (2016) 
found that a high value of exchange rate signals a high level of capital flight.  
 
3.7.1 Variables Used for Regression Analysis 
 Some potential drivers of abnormal pricing include corruption, inflation, Transfer Pricing Risk, 
Transfer Pricing Rank, current account balance, GDP growth, interest rate and the Real Effective 
Exchange Rate (REER). The table below presents the variables together with their sources. 
 
 
84 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 3.3: Variables used for regression analysis 
Variable Source Measurement 
1-100 with higher values showing higher levels 
Trading Economics 
Corruption  
of corruption; Corruption Perception Index 
Trading Economics Measured as a percentage 
Inflation  
Trading Economics Measured as a percentage 
Interest Rate 
Real Effective Exchange Trading Economics Measured as a percentage 
Rate (REER) 
Mescall, & Klassen, Values generated for each country by Mescall 
Transfer Pricing Risk (2018) Adapted from &Klassen (2018); measured as a number 
(TPR) 
Marur (2019) ranging from 0 to 10 
Mescall, & Klassen, Values generated for each country by Mescall 
Transfer Pricing Rank (2018) Adapted from &Klassen (2018); measured in decimal values 
(TPR) 
Marur (2019) ranging from 0 to 1 
Trading Economics Measured in USD 
Current Account Balance 
Trading Economics Measured as a percentage 
GDP Growth 
Estimates Generated  Measured in USD 
Overvaluation (Ove) 
Estimates Generated Measured in USD 
Undervaluation (Und) 
 
 
 
 
85 
 
University of Ghana http://ugspace.ug.edu.gh 
The general form of the model can be specified as: 
Yi,t = α + βXi,t + εi,t 
where the subscript i denotes the cross-sectional dimension and t represents the time series 
dimension. Yi,t represents the dependent variable in the model, while Xi,tcontains the set of 
explanatory variables in the estimation model, α is the constant and β represents the coefficients. 
εi,tis the error term 
The regression model estimated is outlined as follows: 
Undit = α +  β1Infi,t + β2Corrupti,t + β3IntRatei,t + β4TaxDiffi,t + β5Corri,t + β6CABi,t
+ β7GDPGrowthi,t + β8REERi,t + β9TPRi,t + εi,t … … . . (Eqn 13) 
Oveit = α +  β1Infi,t + β2Corrupti,t + β3IntRatei,t + β4TaxDiffi,t + β5Corri,t + β6CABi,t
+ β7GDPGrowthi,t + β8REERi,t + β9TPRi,t + εi,t … … . . (Eqn 14) 
Where Und represents Undervaluation, Ove represents Overvaluation, Inf represents Inflation, 
IntRate represents Interest Rate, TaxDiff represents Tax Differential, Corr represents Corruption, 
CAB represents Current Account Balance, GDP Growth represents GDP Growth, REER 
represents the Real Effective Exchange Rate and TPR represents Transfer Pricing Risk or Transfer 
Pricing Rank.  
 
An Ordinary Least Squares (OLS) and a Random Effects equation are estimated. The Random 
effects model was chosen based on the results of the Hausman test. A Quantile regression is also 
estimated as a robustness check. 
 
 
86 
 
University of Ghana http://ugspace.ug.edu.gh 
3.8 CHAPTER SUMMARY 
 
This chapter outlines the methodology used in the thesis. It explains the qualitative methodology 
used to analyse the commodity value chains and identifies the risks for IFFs and the various 
methods of measuring trade mis-invoicing such as the Partner Country Trade Gap (PCTG) method 
and the Price Filter methods. It also outlines the regression model used to estimate the determinants 
of trade mis-invoicing.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
87 
 
University of Ghana http://ugspace.ug.edu.gh 
CHAPTER FOUR 
DISCUSSION OF RESULTS 
4.1 INTRODUCTION 
This section presents the value chains of the export commodities and the risks for IFFs in the 
commodities. Next, the estimates of illicit financial flows in some of Ghana’s top export 
commodities using the price filter method and the Partner Country Trade Gap (PCTG) method are 
presented. The section goes further to present the estimates and discussion of the determinants of 
trade mis-invoicing. 
 
4.2 VALUE CHAIN ANALYSES AND RISKS FOR IFFS 
The thesis analysed the value chains of the commodities studied: gold, cocoa, bauxite and 
manganese to identify the risks for IFFs along the value chain.  A value chain maps the entire range 
of activities required to bring a product or service through the different phases of production from 
beginning to end (Kaplinsky & Morris, 2003).  
 
Some of the risks identified included: transfer pricing risks due to the nature of ownership of the 
firms operating in the sector; the presence of artisanal, small-scale, and informal firms in the sector; 
regulatory infrastructure for verifying export valuation; and transit trade from neighbouring 
countries 
 
 
 
 
88 
 
University of Ghana http://ugspace.ug.edu.gh 
4.2.1 Gold Value Chain and Risks for IFFs 
Gold is Ghana's most valuable export product, accounting for 49 percent of the country's overall 
value exports (OEC, 2019). In 2017, gold accounted for 96.4% of the total revenues from mineral 
exports from Ghana (Minerals Commission, 2018). In 2018, Ghana became Africa's largest 
producer, producing about 4.8 million ounces of gold, which exceeded South Africa's 4.2 million 
ounces of gold produced (Whitehouse, 2019). Ghana's top four mines for the period understudy, 
have been Goldfields Ghana Limited, Newmont Ghana Gold Limited-Ahafo, Newmont Golden 
Ridge Limited, Akyem, and AngloGold Ashanti Iduapriem Limited in terms of production and 
revenue output according to the Ghana Chamber of Mines 2018 survey (Ghana Chamber of Mines, 
2019). 
 
Overall, the minerals and mining sector is the largest contributor to Ghana’s GDP (Minerals 
Commission, 2017). The fiscal contribution of gold mining to public finances is observed via 
general taxes on profit and labor borne by all companies as well as specific revenue generators 
such as mining royalties, license fees, property rate payments and export duties. Additionally, 
mining companies in Ghana contribute to the socio-economic development of their communities, 
as part of the Corporate Social Responsibility (CSR). According to the Annual Report, 2016 of 
Ghana Chamber of Mines, its producing members invested US$ 12.29 million in their host 
communities. 
 
The Government of Ghana is the overall regulator. Its mandate is exercised through the Ministries 
of Land and Natural Resources (MLNR), the Ministry of Environment, Science, Technology and 
Innovation (MESTI) and the Ministry of Finance and Economic Planning (MoFEP). Other 
89 
 
University of Ghana http://ugspace.ug.edu.gh 
agencies which also have oversight roles include the Minerals Commission, the Lands 
Commission, the Office of the Administrator of Stool Lands (OASL), the Environmental 
Protection Agency (EPA), the Bank of Ghana (BOG), the Ghana Revenue Authority (GRA), the 
One Stop Service Centre (OSSC), the Geological Survey Authority (GSA) and the Precious 
Minerals Marketing Company (PMMC). Others include the town and country planning unit, the 
district mining committees, the security services and the traditional rulers or clans who own lands 
in Ghana.  
 
Traditional Rulers/Clans play a key role in the mining sector. In Ghana, land is vested in the chiefs 
and clan heads and they are mandated to sell or lease to prospective buyers. However, the minerals 
buried in the land are the properties of the state of Ghana according to the 1992 constitution.  
Article 257 (6) of the 1992 Constitution states explicitly that “Every mineral in its natural state in, 
under or upon any land in Ghana, rivers, streams, water courses throughout Ghana and the 
exclusive economic zone is the property of the Republic of Ghana and shall be vested in the 
President on behalf of and in trust for the people of Ghana”. As such, there is sometimes conflict 
between the traditional rulers, new owners of lands and the government in land acquisition, 
exploration and mining. In some instances, traditional rulers have been blamed for the high 
incidence of illegal mining in Ghana.  
 
Gold is mined in almost all the regions of Ghana through Large-Scale Mining (LSM) and Artisanal 
and Small-Scale Mining (ASM). Many of the LSM companies are members of the Ghana Chamber 
of Mines whilst the small-scale miners have their own union known as the Small-Scale Miners 
Association. The large-scale mining companies in Ghana, that are members of the Ghana Chamber 
90 
 
University of Ghana http://ugspace.ug.edu.gh 
of Mines include Gold Fields Ghana Ltd.; AngloGold Ashanti Ltd., Golden Star Resources; and 
Newmont Gold Ghana Ltd. (2016 Annual Report, Minerals Commission). As at 2018, there were 
over 300 ASMs registered in Ghana. Apart from the ASMs, there are also quite a sizeable number 
of illegal miners (galamsey) in the industry. These are managed by foreign operators usually under 
the cover of locals because according to the Minerals and Mining Act (Act 703), small-scale 
mining is a prerogative of only Ghanaians. However, many mining sites are operated by foreigners 
(Hausermann and Ferring, 2018). According to the same authors, this has largely been made 
possible due to loopholes in the acquisition of licenses to operate on small-scale mining concession 
sites (ibid). 
 
In Ghana, the process of extracting gold from its ore is such that mined amalgam gold is smelted 
to obtain more bars that can be up to 23 carats; just below the 24 carats for which prices are usually 
quoted for global trading. After mining, gold is assayed for its value before sale or export. The 
large-scale mining companies do their own assaying at the mines as well as the airports, in the 
presence of a customs officer, before shipment. However, ASMs are required to send their gold to 
PMMC for assaying before shipment. In 2018, a new assaying regime was established in pursuant 
to Regulation (3) of the Minerals and Mining General Regulation 2012 (LI 2173), which requires 
both Small and Large-Scale Mining Companies to assay all their produce at PMMC before export 
from the 19th of February 2018. Due to this directive of PMMC to become the national assayer for 
all, a modern national assay laboratory was built and personnel capacity was upgraded to cater to 
its new role. This is in a bid to have accurate and valid records of all exported gold values. PMMC 
has also agreed with GRA (Customs Division) that allows gold produced in neighbouring countries 
to be brought into Ghana for assaying, documentation and shipment as Transit Gold. 
91 
 
University of Ghana http://ugspace.ug.edu.gh 
Most of the gold ore bars mined in Ghana are exported while the rest are mainly made into jewelry 
and sold on the local market. Ghana exports gold to Switzerland, South Africa, Netherlands, 
Pakistan, Sierra Leone, India, Lebanon, Saudi Arabia, United Arab Emirates, Turkey, Singapore 
and Hong Kong. The types of gold exported include semi-processed unwrought gold, monetary 
gold, other semi-manufactured forms of gold, powder of gold and non-monetary gold. Unlike 
cocoa beans that is exported by only one institution in Ghana, there are many licensed gold 
exporters, including AA Minerals and Adfat Company Limited, to mention a few. Until 2013, 
there were no gold refineries in Ghana. Asap Vasa Gold Refinery, Sahara Royal Gold Refinery, 
and Gold Coast Refinery are recent firms that were established in 2013, 2015, and 2017 
respectively. 
 
The risks for IFFs in the gold sector are identified as follows: 
Transfer pricing risks arising from multinational firms’ international operations: Many 
leading mining firms in Ghana are multinational corporations with offices outside Ghana and 
several other foreign branches in various tax and legal jurisdictions (Oppong, 2013). This leads to 
substantial economic incentives for transfer mispricing and tax optimization through intra-
company exchange and financial transfers. Readhead (2016) contends that the extractive sector to 
Ghana's economic growth could be much higher, except for tax evasion by companies operating 
in the sector by mispricing, mispricing of trade and thin capitalisation. In order to minimise their 
state tax liability, the sector-based MNCs may have an incentive  to under-report their exports. The 
explanation is that the revenue earned by the state from the industry primarily comes from 
corporate and income taxes, royalties as well as withholding tax (Minerals Commission, 2018; 
Personal Communications, A. Tawiah; Aryee, 2018). 
92 
 
University of Ghana http://ugspace.ug.edu.gh 
Artisanal, small-scale, and informal mining firms: Artisanal, small-scale, and informal mining 
is a characteristic of Ghana's intensively worked local output of gold mineral and ore bars by small-
scale concession holders (CSSM, 2019). Gold mining is permissible for people, small groups (less 
than nine persons), or a cooperative society with ten or more members (Parliament of Ghana, 
1989). The mining of small mines is a prerogative of Ghanaians according to the Minerals and 
Mining Act of Ghana (Act 703), which is popularly called galamsey. The lack of domestic capital 
and technological support from the government, however, encourages foreign enterprises to 
capture small-scale mining concessions through local or traditional leaders. The role of these 
foreign actors will extend to the entire activity of a mining site beyond providing technical support 
(Hausermann & Ferring, 2018). This sector is difficult to regulate due to its small-scale, 
geographically scattered and informal existence involving numerous intermediaries. Licensed gold 
exporters can also buy gold from licensed and non-licensed gold miners in Ghana. This situation 
allows intermediate buyers and other persons to export undervalued and/or misreported gold doré 
bars and engage in trade-based money laundering (Hunter 2020; Personal Communications, K. 
Opare-Hammond). 
 
Regulatory infrastructure for export valuation: Before shipment, the gold bars must be assayed 
to assess the purity of their contents. Large mining companies typically perform their own assay 
or use foreign pre-shipment inspection firms to certify the contents of each shipment 
independently. Dedicated customs officers are supposed to be present at the stage where gold bars 
are assayed but may neither be present nor qualified to decide how gold is assayed. 
 
93 
 
University of Ghana http://ugspace.ug.edu.gh 
The verification of the assay values is only conducted at a foreign refinery after the gold has 
been exported (Personal Communications, C. Nyarko). Any value anomalies found would likely 
be disclosed only to the mining enterprise and not to the state. This can contribute to revenue losses 
for the state if the business declares the prices of the export of gold ore bars. In addition, artisans, 
and small and informal mining companies that lack their own assays must submit their gold to a 
state-owned company, Precious Minerals Marketing Company (PMMC), for assay before 
shipping. 
 
However, until 2016, due to this lag in gold valuation, the coverage of the testing infrastructure 
continued to be minimal. Under the new legislation, the Minerals and Mining Act of Ghana (Act 
703), small and large mining firms are now obliged to test the government regulator, PMMC, for 
their golden shipments. A new national assay laboratory was therefore designed and fitted with 
professional staff (Minerals Commission, 2018; Personal Communications, K. Opare-Hammond). 
 
Transit trade from neighboring countries: Trade flows can be listed as exports, imports, transit, 
and re-exports (Bensassi & Jarreau, 2019). The government of Ghana allows for gold mined in 
neighboring countries to be assayed, documented, and exported from Ghana as transit gold. 
However, this creates a situation where local actors can corrupt regulators and combine gold from 
different sources to misreport its true value and origin before shipping abroad (Rahman, 2018). 
The Head of the Tax Policy Unit at the Ministry of Finance reported that Ghana Revenue Authority 
(GRA) had lost an accumulated revenue of GH¢4.5 billion over the past five years as a result of 
infractions by traders, and importers and governmental agencies (Ocloo, 2018). The government 
94 
 
University of Ghana http://ugspace.ug.edu.gh 
of Ghana lost GH¢2 billion in revenue in 2018 through non-observance of rules governing transit 
trade and the probable complicity and collusion of Customs officials (Sarpong, 2019).  
  
4.2.2 Cocoa Sector 
Ghana is the second largest cocoa bean producer in the world. Cocoa exports account for around 
one-quarter of the total export profits (Peprah, 2019). The sector employs nearly three million 
farmers, provides business to service providers using shipping, warehousing, insurance and so on; 
and contributes by bursaries to primarily farmers' wards and the needy people in cocoa farming 
communities. Cocoa is a primary source of livelihood for many rural communities in the southern 
part of the country, where it is largely grown. This sector accounted for approximately 2.3% of 
Ghana’s annual gross domestic product and 9.9% of agricultural output between 2010 and 2017.  
 
Cocoa is the most important agricultural export commodity, contributing about 80.7% of 
agricultural export averages between 2014 and 2017 (ISSER, 2018).  Due to the socio-economic 
value of the commodity, the Ghana Cocoa Board (COCOBOD) regulates the industry with great 
interest. Around 70%-80% of the produced cocoa is exported (mainly the bigger beans from the 
main cultivation for the year). The remaining beans, mostly the small beans, are primarily sold to 
local manufacturers, i.e., both the Cocoa Processing Company and others in producing cocoa 
products such as cocoa pastes, liquors, powder, husks, and other products such as cocoa waste.  
 
Ghana’s specialised cocoa business regime is such that the Government of Ghana (GoG) plays a 
dominant role in a partly liberalised market. This has been described as unique in the industry 
(Williams 2009). GoG, through Ghana Cocoa Board (COCOBOD), regulates the sector with keen 
95 
 
University of Ghana http://ugspace.ug.edu.gh 
interest due to the revenue generating and socio-economic benefits derived from the commodity.  
Other statutory bodies with some oversight for the sector include the Ministry of Finance and 
Economic Planning (MoFEP), the Ministry of Food and Agriculture (MoFA), and the Ministry of 
Lands and Natural Resources (MLNR). However, it has been observed that the various regulatory 
agencies do not co-operate fully. This has been identified as a disadvantage to the reaping of the 
full potential benefits of the commodity.  
 
COCOBOD, the main link between the Government of Ghana (GoG) and the cocoa sector, was 
established by ordinance in 1947. Its mission is to encourage and facilitate the production, 
processing, and marketing of cocoa efficiently and cost-effectively. To achieve its mission, 
COCOBOD has several subsidiaries which enable it to be engaged in the cocoa sector at all levels. 
These subsidiaries are the Seed Production Division (SPD), the Cocoa Research Institute of Ghana 
(CRIG), the Cocoa Health and Extension Division (CHED), the Quality Control Company (QCC) 
and Cocoa Marketing Company (CMC).  
 
After production at the farm level, the cocoa beans are dried, examined by the Quality Control 
Company (QCC) and then sold to Licensed Buying Companies (LBCs). The LBCs sector is fully 
liberalised; this has led to some cocoa processing companies owning LBCs. A farm-gate price is 
set by the Producer Price Review Committee (PPRC) at the beginning of the farming season. This 
guarantees the fair minimum price to be paid to the farmers. COCOBOD requires more than $1 
billion US dollars annually to finance its crop purchases. COCOBOD obtains an off-shore trade 
finance facility via syndicated loans from a consortium of international and local banks to enable 
it to raise funds to pay cocoa farmers via LBCs. After purchase, the beans are transported to various 
96 
 
University of Ghana http://ugspace.ug.edu.gh 
ports for export. The Cocoa Marketing Company (CMC) is solely mandated to sell and export 
cocoa beans in Ghana. Its trading unit sells the cocoa beans, while the port operations and shipping 
units are responsible for transporting and exporting the beans.  
 
About 80% of the cocoa (mostly the big main crop beans and some light crops) is exported. For 
the period 2011 - 2017, Ghana exported a total of 5.5 million metric tonnes of cocoa to countries 
such as Turkey, Brazil, Malaysia, Belgium, the United States, Japan, Great Britain, Germany, Iran, 
Netherlands, Switzerland, and New Zealand, and Ireland among others. Ghana’s foremost cocoa 
trading partner is the Netherlands. Chocolate and other confectionary items are also made for sale 
both locally and in international markets. The cocoa processing companies in Ghana include local 
companies such as Cocoa Processing Company (CPC), Plot Enterprises, and multinational 
companies such as Cargill, Barry Callebaut, and others. 
 
Risks for Trade Mis-invoicing 
Transfer pricing risks due to multinational firms’ international trade operations: Transfer 
mispricing in the cocoa bean industry is not a major risk in Ghana, because COCOBOD in the 
government of Ghana dominates the market and serves as a regulatory and supervisory body to all 
operations in the market. The operations of state-owned companies permeate all stages of cocoa 
bean production. Some of the key acting stakeholders include the Ghana Cocoa Research and 
Institute (GRIG), the SPD, seeds supplying farmers with seeds, the Divisional Health and 
Extension of Community (CHED), the farmers supporting service, the Quality Control Division 
(QCD), which certifies the value of cocoa beans, and the Cocoa Mar research division, which is 
the most important contribution to our report (David 2013; Personal Communications, COCOBOD 
97 
 
University of Ghana http://ugspace.ug.edu.gh 
Personnel). Since all but one of the COCOBOD subsidiaries is a state-owned corporation, it is 
presumed that most cocoa export revenues return to the state to establish a situation where transfer 
errors are not a major problem (Bulir, 1998). In addition, an audit committee has been formed 
since 2018 to oversee the administration of the Board's financial resources following the 
regulations that oversee the financial management in government institutions. This is following 
the 2016 Public Financial Administration Act (Ampofo, 2018). The cocoa paste industry does have 
several multinational companies (MNCs), however, that provide some transfer mispricing 
incentives in Ghana to lower the tax liabilities (Kwaramba et al., 2016). 
 
Artisanal, small-scale, and informal firms:  In cocoa production and exports, some peculiarities 
could render unto the risks of mis-invoicing. This study finds that the government finds it difficult 
to control artisanal, small-scale, and informal companies operating in the cocoa paste industry. 
This lack of control and regulatory adherence can lead to some risks of evasion and non-
compliance in the field. Consequently, this could and has potentially resulted in Ghana losing 
substantial tax revenues. For instance, while Ghana Investment Promotion Council (GIPC) 
includes multinationals operating in the region, the same criteria do not apply to craft, small and 
informal companies (Personal communications, M. Acheampong). Moreover, COCOBOD and 
Ghana Export Promotion Authority (GEPA) partially control the cocoa paste market (Personal 
communications, E. Quao). This creates loopholes within the sector those companies can take 
advantage of to reduce their tax liability to the government, particularly when the state institutions 
that oversee the sector do not have effective cooperation and information sharing systems or 
procedures. 
98 
 
University of Ghana http://ugspace.ug.edu.gh 
Regulatory infrastructure for verifying export valuation: The COCOBOD Quality Control 
Division verifies before the sale of both the cocoa beans and paste for export. The main crop, cocoa 
beans, are exported by CMC directly. In contrast,the light crop cocoa beans are processed for 
export and domestic consumption by cocoa paste companies (Personal communications, CMC 
Personnel). Approximately 90 per cent of all cocoa processed is exported and the remaining 10 
per cent is used to produce confectionary (Ashitey, 2012). Though CMC sells cocoa paste 
products, supervision is not as strict as it is with cocoa beans since they are not owned directly by 
COCOBOD. 
 
Furthermore, since many companies in the sector are multinationals, they are regulated by the Free 
Zone Authority of Ghana (Personal communications, M. Acheampong; Personal communications, 
F. Mate-Kodjo). This creates a situation in which several entities supervise the same sector and 
thereby create some differences and a lack of consensus about data and regulated procedures within 
the sector. The multinationals in the sector, have the potential to minimise their tax burden on the 
State. 
 
Transit trade from neighboring countries: Like gold, there are risks to the trading of cocoa 
products from neighbouring countries. Several cases of cocoa beans between Cote d'Ivoire and 
Ghana have been reported based on the price differentials between the two countries (Bulir, 1998). 
Reuters estimated that in 2017, approximately 100,000 to 200,000 tonnes, because of the higher 
prices Ghana is giving farmers exports to Ghana from nearby Ivory Coast (Kpodo, 2017). The two 
countries have, however, agreed since 2018 to announce their prices jointly to end the phenomenon 
(Bruce, 2018). 
99 
 
University of Ghana http://ugspace.ug.edu.gh 
4.2.3 Bauxite 
Bauxite is an ore which serves as the world’s leading source of aluminum. It was first discovered 
by a French Geologist named Pierre Berthier in 1821 in France. Currently, Australia is the world’s 
largest producer of bauxite, followed by China. Other leading bauxite producers include Guinea, 
Brazil, Canada, India, and Russia. Bauxite is usually processed first into alumina and then to 
aluminum. For bauxite to be commercially exploitable, it should contain more than 25–30% 
aluminum oxide. The major steps in aluminum processing are bauxite mining, refining of bauxite 
into alumina, smelting of alumina into raw aluminum ingots, and processing or casting of 
aluminum ingots into sheets, coils, and profiles. The processed aluminum is the input for aluminum 
goods ranging from packaging material, automotive parts, and construction elements to household 
goods.  Although most of the world’s bauxite reserves are found in tropical and subtropical regions, 
converting aluminum from bauxite requires a lot of investment and advanced technology, 
including cheap electricity. As such, aluminium production tends to be concentrated in developed 
countries.  
 
The African history of bauxite and aluminum is quite recent and started in the early 1950s when 
demand exploded, partly because of the Korean War and, consequently, when European and 
North-American companies or consortiums began to prospect Guinea, Ghana, or Cameroon for 
investment (Lesclous, 2002). Although Alcan had mining concessions on the Loss Islands in the 
early 1920s, the Guinean saga for this Canadian group really began with the discovery of the 
renowned Boke ore field in 1952. From that time on, Guinean willingness to process bauxite was 
explicit but was, with rare exceptions, constantly rebuffed by the multinationals operating in the 
country (Lesclous, 2002). 
100 
 
University of Ghana http://ugspace.ug.edu.gh 
In Ghana, the mining sector includes 16 large-scale mining companies, of which 14 produce gold, 
one produces bauxite, and the other produces manganese. Also, over 300 registered small-scale 
mining groups and 90 mine support service companies (GIPC, 2016).  
 
Bauxite was first discovered in Ghana in 1914, but exploration and mining works started in the 
1940s. Currently, Ghana is the 17th largest exporter of bauxite in the world (Index Mundi) and has 
some of the world’s largest bauxite reserves, with 960m tonnes worth an estimated $460bn. 
Bauxite can be found in many places in Ghana, such as the Sefwi Bekwai area in the Western 
Region, which the Ghana Bauxite Company is mining with available reserves of 29.5 Mt, sufficient 
to support a mine life of 100 years. Other deposits include those at Nyinahin near Kumasi and in 
the Atewa range. The Nyinahin deposit has a total inventory of 350 metric tonnes (MT) at an 
average grade of 48.9% to 51% alumina and 2.8% to 4.4% silica. 
 
In contrast, that of the Atewa range has about 150 MT ranging in grade from 40.8% to 45.7% 
alumina and 1.8% to 3.9% silica (Kesse, 1985). Ghana’s bauxite is described as high alumina and 
low silica bauxite ore, which is among the best deposits of bauxite. There are several types of 
bauxite depending upon the end use. Some examples include the metallurgical-grade bauxite 
which is used for aluminum production and the refractory-grade bauxite which is used in the 
ceramic and refractory industries. The grade of alumina in bauxite ores in Ghana falls below the 
refractory grade. 
 
 In 2016 alone, 1.278 million metric tonnes of bauxite were produced in Ghana, while a total of 
1.476 million metric tonnes of bauxite was produced in 2017, amounting to an increase in bauxite 
101 
 
University of Ghana http://ugspace.ug.edu.gh 
exports from $38.7m in 2016 to $50.9m. The quantity, however, declined to 1.011 million metric 
tonnes in 2018 (Minerals Commission, 2017; 2018).  
  
The Government of Ghana regulates the Bauxite industry through the Ministry of Lands and 
Natural Resources and its relevant agencies, such as the Minerals Commission, the Geological 
Survey Authority, and the Environmental Protection Agency (EPA).  
 
Historically, the sector has been centred on one company, the Ghana Bauxite Company, and its 
mine at Awaso.  The company is currently owned by the Bosai Group, a Chinese-owned company 
that acquired it in 2009 from Rio Tinto Group. The Bosai Group is a mining company specialising 
in bauxite and manganese and can be found in Chongqing, Southwest China. It also engages in 
aluminum and ferroalloy production. 
 
The mine at Awaso is an open pit with a low stripping ratio, which poses little difficulty for mining 
operations. Exploration is the first step in the extraction process, which is followed by mining and 
transporting the extracted materials to the export port. After the exportation of bauxite to smelters 
in Scotland and Canada, alumina is imported from Jamaica and the United States of America for 
the local smelter; Volta Aluminium Company Limited (VALCO) which has faced significant 
challenges in recent years, particularly energy issues (Oxford Business Group, 2013).  In 1965, the 
Volta River Project led to the creation of the Akosombo dam, which was to provide electricity for 
the Volta Aluminum Company (VALCO), which, in turn, was expected to provide enough revenue 
to mine the local bauxite resources in Aya-Yenuhin and end alumina imports. 
 
102 
 
University of Ghana http://ugspace.ug.edu.gh 
Additionally, the Government of Ghana announced its intention to develop the country’s bauxite 
value chain and address the factors impeding the industry’s efficiency and cost-effectiveness. 
Interactions with relevant players and stakeholders in the Bauxite industry indicate that there are 
potential for downstream processing of aluminum into different aluminum products. These 
products in turn could serve as inputs to Ghanaian firms, such as construction and other aluminium-
oriented industries. Such aims in higher value aluminum products are quite ambitious. They would 
require overcoming the energy as mentioned above and capacity constraints and an increase in 
production to achieve economies of scale.  
 
The Government of Ghana as part of its efforts to establish an aluminum industry created the 
Ghana Integrated Aluminum Development Corporation in 2018. The roles of the corporation 
include promotion of responsible mining and ensuring the development of the industry. They are 
also responsible for developing necessary infrastructure, including rails, roads and energy, 
industrial parks and associated social infrastructure to support related businesses in the sector and 
the industry as a whole. 
 
So far, the corporation has identified some bauxite deposits that could form the basis of new mining 
operations, including a deposit of around 150 million metric tonnes of the mineral underneath the 
Atewa Forest, one of the last enduring rainforests in Ghana. However, there has been a backlash 
about the potential damage to the environment. 
Risk for trade mis-invoicing 
Transfer pricing risks due to multinational firms’ international trade operations: Trade mis-
invoicing due to multinational firms’ international trade operations is a risk in Ghana’s Bauxite 
103 
 
University of Ghana http://ugspace.ug.edu.gh 
sector since a foreign company owns the Ghana Bauxite Company Limited; a Chinese Company. 
There is a high risk of export undervaluation for developing countries, and in this instance Ghana, 
because the amount of tax these companies pay is usually determined by how much has been 
exported. According to Readhead (2018), commodities such as Bauxite are vulnerable to 
undervaluation due to a lack of quoted bauxite prices and limited spot sales. In 2018, Ghana 
Bauxite Company was slapped with a GHC 3.5 million tax liability for transfer pricing and 
underselling bauxite to its mother company. 
 
4.2.4 Manganese Value Chain and Risks for IFFs 
Manganese is a transition metal that is a vital industrial alloy. It is used in iron and steel production, 
manufacturing of dry cell batteries, fuel oil additives, anti-knock agents, manufacturing of glass, 
the production of potassium permanganate and the production of fungicides and tanning of leather.  
However, despite its general usefulness, it is known to be a common environmental contaminant, 
which can cause toxic effects in humans (Milatovic et al. 2011). Some of the top countries where 
manganese resources are found include South Africa, Australia, China, Gabon, Brazil, India, 
Kazakhstan, Ghana, Ukraine and Malaysia. About 80% of the world’s manganese resources are 
and recognised in South Africa. 
 
The manganese value chain consists of three segments: ore producers, alloy producers, and steel 
producers. Ore characteristics segment the manganese ore industry. It comprises of high-grade ore 
(with more than 35% manganese contents) producers who account for two-thirds of production, 
and low-grade ore producers contributing a third of production. Alloy production is performed 
either by independent alloy smelters (70% of production) or by integrated alloy smelters (30% of 
104 
 
University of Ghana http://ugspace.ug.edu.gh 
production). The latter are vertically integrated firms involved in both ore mining and smelting. 
The end-use customers are primarily steel producers (94% of demand) who can be classified into 
integrated mills, mini mill flat producers, mini mill long producers and speciality mills. Chemical 
and specialist metallurgical segments contribute to the balance of the demand. 
 
In Ghana, manganese was first discovered by a colonial official, Albert Ernest Kitson in 1915 and 
mining works began in 1916. The Government of Ghana is the overall regulator of this sector and 
directives can be expressed through the Lands and Natural Resources Ministry. Various agencies 
under the Ministry, such as the Minerals Commission and the Geological Survey Authority also 
have oversight of the sector.  
 
Currently, two manganese deposits can be found in the Yakaw area near Dixcove in the Western 
Region of Ghana. These have estimated reserves of about 5.5 MT with 42% manganese content 
and 1.0 MT containing 32% manganese content, respectively. Two other deposits can be found in 
the Northern Region with total reserves of over 13.8 Mt with a manganese content of 30% to 45%. 
However, the only operating manganese mine in Ghana is at Nsuta near Tarkwa with 37 Mt of 
manganese ore reserves (Kesse, 1985). Since mining activities began in 1916, the mine has 
produced and exported about 28 MT of high-grade manganese oxide ore. The types of manganese 
ore exported include the battery grade (Mn > 52%), metallurgical grade (47 – 52% Mn), and 
chemical grade (40 - 46% Mn). With the oxide ores almost depleted, the mine has begun mining 
and processing the low-grade carbonate ores which are available in large quantities on the mine. 
The average manganese content of the carbonate ore is 34.5% (Amankwah et al., 1999). 
 
105 
 
University of Ghana http://ugspace.ug.edu.gh 
Manganese in Ghana is mined solely by Ghana Manganese Company Limited (GMCL), one of 
the world's leading producers of manganese. It operates the largest manganese reservoir in Ghana 
in the Nsuta Manganese mine. There are also proprietary mining rights in an area 175 kilometres 
refrom Nsuta and around. The headquarters of GMCL is in the Western Region of Ghana.  
 
GMCL’s Nsuta mine has considerable potential. At the same time, mining operations on the site 
date back to 1916, less than 3% of the 175-sq-km concession has been mined. The concession has 
45m tonnes of proven reserves and the ores have one of the highest manganese-to-iron ratios in 
the world with a low level of impurities, which means that they are well-suited for alloy and 
manganese metal production. The qualities of the Nsuta reserves’ carbonate ore was one of the 
most important factors behind TMI’s acquisition of Consmin. Its immediate expansion potential 
is set to meet the Chinese company’s growing feedstock needs, as one of the world’s leading 
producers of electrolytic manganese metal. 
 
Manganese production in Ghana increased from 2.034 million metric tonnes in 2016 to 3.003 
million metric tonnes in 2017 to 4.552 million metric tonnes in 2018 (Ghana Chamber of Mines, 
2018; 2019). The substantial increase in manganese production in 2017 was largely attributable to 
rising global demand. The sector was able to increase output as a result of substantial capital 
investments made by the Ghana Manganese Company (GMCL) after its Australia-based parent 
company, Consolidated Minerals (Consmin), was acquired by a subsidiary of China's Ningxia 
Tianyuan Manganese Industry (TMI) in a May 2017 deal. In 2017, the Ghana Revenue Authority 
(GRA) recognised Ghana Manganese Company as the third-largest taxpayer (Ghana Chamber of 
Mines, 2018). 
106 
 
University of Ghana http://ugspace.ug.edu.gh 
Risk for trade mis-invoicing 
Transfer pricing risks due to multinational firms’ international trade operations: The 
company's foreign ownership presents a risk for IFFs because of the transfer pricing risks 
associated with foreign ownership of firms. Consmin (Consolidated Minerals), a Jersey-based 
company owns 90% of the company, while the government of Ghana owns 10%. In 2017, Consmin 
was acquired by China Tian Yuan Manganese Limited, a subsidiary of Ningxia Tian Yuan Industry 
Co., Ltd. The comparison of Ghana’s exports of Manganese with China’s imports of Managanese 
show that the reported exports from Ghana are only half of the reported imports of China which 
could be indicative of trade mis-invoicing. 
 
4.3 ESTIMATES OF TRADE MIS-INVOICING 
This section presents the estimates of trade mis-invoicing for the selected gold, cocoa beans, cocoa 
paste, bauxite, and manganese exports using the methodologies and data explained earlier. The 
analyses presented in this section are based on export trade between Ghana and various partners, 
hence, the estimates are general rather than specific to a particular trading partner. Overall, the 
estimates show an economically significant undervaluation of some of the commodities. 
 
 
 
 
 
107 
 
University of Ghana http://ugspace.ug.edu.gh 
4.3.1 Estimates of Trade Mis-invoicing in Gold (HS 7108.13.1000&7108.12.0000) 
Table 4:1 Undervalued and Overvalued Exports-Gold (HS 7108.13.1000&7108.12.0000) 
Year Free-market Price Free-market Price Partner Country 
Filter: Minus 10% of Filter: Plus 10% of Trade Gaps (USD, 
Benchmark Prices* Benchmark Prices* Million) 
(USD, Million) (USD, Million) 
2011 222.8 2.0 -5,145 
2012 544.5 14.5 -4,115 
2013 514.7 23.1 -2,010 
2014 205.5 13.4 -1,447 
2015 78.9 7.4 142 
2016 1,195.4 1.9 909 
2017 1,067.4 4.2 -1,312 
Mean 547.0 9.5 -1,854 
Total 3,829.1 66.6 -12,977 
Data Source: Ghana Revenue Authority (GRA); Metal Focus Limited; United Nations Comtrade Database 
(accessed:December 2020) *The benchmark prices are estimated based on the heterogeneity nature of gold doré 
exported which are gold – silver splits; 100 - 0%, 96 – 4%, 86 – 14%, 81 – 19%, and 67 – 33%. 
Notes: Gold is semi-manufactured; a combination of two export types mis-classified in the GRA data as gold 
bullion (HS:7108.13.1000) and unwrought gold, non-monetary including gold plated with platinum (HS: 
7108.12.0000). Free-market reference price is the daily gold bullion price from London Bullion Market Association 
(LBMA) ($/t oz).  
Negative numbers indicate over-invoicing while positive numbers indicate under-invoicing for the PCTG 
methodology. 
 
Estimates of mis-invoiced exports: Using the lower bound of the arm’s length price range as 
described in the methodology, it is estimated that USD 3.8 billion exports are undervalued. This 
constitutes approximately 11% of the total value of gold exported (USD 35.6 billion). The top 
five destination countries for these undervalued gold exports are India (USD 2.0 billion), South 
Africa (USD 659.3 million), United Arab Emirates (USD 457.7 million), Switzerland (USD 
343.4 million) and Portugal (USD 133.5 million). Their respective percentage share of the total 
value of gold exported within the study period is approximately 5.5%, 1.9%, 1.3%, 1.0% and 
108 
 
University of Ghana http://ugspace.ug.edu.gh 
0.4%. Finally, the total estimated tax base erosion from Ghana due to the undervaluation of gold 
exports is USD 957.3 million. 
Estimates of Overvalued Exports: Similarly, using the upper bound arm’s length range, it is 
estimated that total overvaluations of gold exports between 2011 and 2017 equal USD 66.6 
million. Relative to the total value of gold exports, overvaluation seems negligible, at 
approximately 0.2%. Top five destination countries that overvalued Ghana’s gold doré exports are 
South Africa (USD 31.8 million), Switzerland (USD 22.7 million), The Netherlands (USD 9.7 
million), United Arab Emirates (USD 1.4 million) and India (USD 0.8 million). The over-
valuation could be differences in the assay values of gold exported. 
 
Comparison with Trade Gap Estimates for Partner Countries: In addition, the more 
prevalent partner-country trade gaps are calculated based on asymmetries between Ghana's 
exports and the imports reported by its trading partners using annual product-level data from 
the UN COMTRADE database. The overall results indicate that reported Ghanaian exports 
exceed what the rest of the world reported as imports from Ghana between 2011 to 2017, i.e. 
export over-valuation, which is contrary to the price filter estimates for the same period. These 
conflicting results are primarily driven by missing imports data in the UN COMTRADE database 
from significant destinations for Ghanaian gold, including South Africa, Switzerland and United 
Arab Emirates.  For example: according to the UN COMTRADE database, South Africa 
reported no imports of gold from Ghana between 2011 and 2017. However, in the data provided 
by Ghana Customs, it is reported that Ghana exported USD 4.4 billion of gold to South Africa 
during this period. 
 
109 
 
University of Ghana http://ugspace.ug.edu.gh 
4.3.2 Estimates of Trade Mis-invoicing in Cocoa Beans (HS: 1801.00.1100) 
 
Table 4:2 gives the estimated magnitudes of undervaluation and overvaluation of cocoa beans 
for the period 2011 – 2017. 
Table 4.2: Undervalued and Overvalued Exports – Cocoa Beans (HS: 1801.00.1100) 
Year Free-market Price Free-market Price Partner Country 
Filter: Minus 20% Filter: Plus 30% (USD, Trade 
(USD, Million) Million) Gaps (USD, 
Million) 
2011 24.4 3.8 503 
2012 8.5 7.9 339 
2013 1.4 21.6 361 
2014 57.4 12.0 139 
2015 2.2 7.0 -915 
2016 9.0 47.0 262 
2017 23.8 15.4 13 
Mean 18.1 16.4 100 
Total 126.6 114.7 703 
Data Source: Ghana Revenue Authority (GRA); United Nations Comtrade Database (accessed: December 2020) 
Notes: Cocoa beans are superior quality raw beans (HS: 1801.00.1100). Free-market price is the London Futures 
Price (US$/MT). 
Negative numbers indicate over-invoicing while positive numbers indicate under-invoicing for the PCTG 
methodology. 
 
 
 
110 
 
University of Ghana http://ugspace.ug.edu.gh 
Estimates of undervalued exports: Despite the unique marketing characteristics of Ghana 
cocoa beans exports, it is estimated that about USD 126.6 million worth of cocoa beans was 
undervalued between 2011 and 2017 (Table 4:2). This equals about 1.0% of the total value of 
cocoa beans exported (USD 12.6 billion) which appears to be a small proportion but equals an 
economically significant gross amount. The top five countries that undervalued cocoa beans 
exported from Ghana are Estonia (USD 45.7 million), The Netherlands (USD 14.7 million), 
Germany (USD 9.7 million), China (USD 9.5 million), and Belgium (USD 8.8 million). Their 
corresponding percentage share of total cocoa beans exports are 0.4%, 0.1%, 0.08%, 0.08% and 
0.07%. Although undervaluation in percentages appears small, revenue loss is not negligible. 
There is an estimated tax base erosion of USD 31.6 million. The under and over-valuation 
observed could be as a result of the methodology used. 
 
Estimates of overvalued exports: The overvalued cocoa beans exports are quite similar to the 
undervaluation estimates. For example, the estimated overvalued amount of cocoa beans is USD 
114.7 million, or 0.9% of total exports over the entire study period (Table 5). The top five 
destination countries that overvalued cocoa beans from Ghana are Malaysia (USD 26.0 million), 
The Netherlands (USD 18.7 million), Brazil (USD 15.9 million), and the United States of 
America (USD 11.5 million), and Japan (USD 7.0 million). 
 
Comparison with partner country trade gap estimates: In addition, an estimate of partner-
country trade gaps between Ghana’s reported exports of cocoa beans and its trading partners’ 
reported imports from Ghana is generated. Overall, the findings indicate that reported Ghanaian 
exports are significantly lower than the reported imports by the rest of the world from Ghana, 
i.e. export under-valuation. These estimates of undervaluation are significantly higher than what 
111 
 
University of Ghana http://ugspace.ug.edu.gh 
was estimated using transaction-level export data. The hypothesis is that these differences are 
primarily driven by the entrepot trade effect in international trade statistics, whereby the exporting 
country records an intermediate transit country or shipping hub as the final destination of their 
exports. However, the products are re-exported from the transit country to the final destination, 
which reports them as imports from the source country. This is a significant feature of the 
international trade in cocoa beans, which are usually purchased and stored for multiple years 
before being shipped to their final destination. 
 
4.3.3 Estimates of Trade Mis-invoicing in Cocoa Paste (HS: 1803.20.0000). 
The annual and total undervaluation and overvaluation estimates of cocoa paste are given in 
Table 4:3. 
 
Estimates for undervalued exports: The estimates indicate that the amount of undervalued 
cocoa paste is USD 130.5 million (or 7.2% of the total export value) between 2011 and 2017. 
Although endogenously determined, the interquartile range filter provides better abnormal 
pricing estimates. The lack of benchmark prices makes it unfeasible to estimate the arm’s length 
price range using commodity market prices. This undervaluation could be attributed to the 
different cocoa paste/mass products combined to generate the estimates (COCOBOD Personnel 
Feedback, Stakeholder Meeting, 2019). 
 
Another reason is the high number of multinationals operating in the sector, which increases risks 
for trade and transfer mispricing in the sector (Kwaramba et al., 2016). The top five countries that 
undervalued cocoa paste from Ghana are Spain (USD 46.2 million), Bulgaria (USD 36.9 million), 
The Netherlands (USD 13.5 million), Turkey (USD 5.4 million) and Russia (USD 3.3 million); 
112 
 
University of Ghana http://ugspace.ug.edu.gh 
with a percentage share of total exports as 2.5%, 2.0%, 0.7%, 0.3% and 0.2% respectively. The 
total tax base erosion due to the overall undervaluation estimated is USD 32.6 million. 
 
Estimates for undervalued exports: The overvaluation of cocoa paste amounts to approximately 
USD 59,1 million, or approximately 3.2% of the total export value of cocoa paste. The first five 
countries that overvalued cocoa paste are The Netherlands (USD 13.2 million), Turkey (USD 8.3 
million), Bulgaria (USD 5.9 million), United States of America (USD 4.6 million) and United 
Kingdom (USD 4.0 million). 
 
Comparison with partner country trade gap estimates: In addition, partner-country trade gaps 
estimations are done for cocoa paste exports from Ghana and the results are reported in Table 
4:3. Significant data misreporting is recorded in product-level trade statistics for cocoa paste by 
Ghana Customs which does not allow for reliable interpretation of the observed trade gaps. 
Transaction-level data from Ghana Customs shows that all export transactions for cocoa paste, 
cocoa butter and cocoa powder have been recorded under the tariff category of cocoa paste despite 
being allocated to distinct tariff categories. This leads to incomparability with the import statistics 
of its trading partners. Overall, this case highlights the strong limitations of estimating trade 
mispricing using aggregate trade data in the case of developing countries which limited statistical 
capacity and misreporting. 
 
 
 
113 
 
University of Ghana http://ugspace.ug.edu.gh 
 
Table 4.3: Undervalued and Overvalued Exports – Cocoa Paste (HS:1803.20.0000) 
Year Rolling Interquartile Rolling Interquartile Partner Country 
Range Filter: Below 25th Range Filter: Above Trade Gaps 
pctle (USD, Million) 75th pctle (USD) (USD, Million) 
2011 2.7 0.2 48 
2012 7.7 4.1 - 
2013 11.8 4.5 26 
2014 33.7 19.1 -401 
2015 24.9 14.9 -393 
2016 18.4 7.7 - 
2017 31.3 8.6 -258 
Mean 18.6 8.4 -195 
Total 130.5 59.1 -977 
Data Source: Ghana Revenue Authority (GRA); United Nations Comtrade Database (accessed: December 2020) 
Notes: Cocoa paste includes wholly and partly defatted (HS: 1803.20.0000).  
Rolling Interquartile range is calculated for unit prices (USD per kg) using transaction-level data from GRA. 
Negative numbers indicate over-invoicing while positive numbers indicate under-invoicing for the PCTG 
methodology. 
 
 
 
 
 
114 
 
University of Ghana http://ugspace.ug.edu.gh 
4.3.4 Estimates of Trade Mis-invoicing in Bauxite (HS:2606.00.0000) 
 
Table 4:4 Undervalued and Overvalued Exports – Bauxite (HS:2606.00.0000) 
Year  Interquartile Interquartile Partner Country 
Range Filter: Below Range Filter: Above Trade Gaps (USD, 
25th pctle (USD, 75th pctle (USD, Million) 
Million) Million) 
2011 0.03 1.81 7 
2012 1.17 0.72 -22 
2013 0.27 2.81 21 
2014 17.15 1.11 32 
2015 22.15 0.19 23 
2016 0.01 3.77 41 
2017 - - 55 
Mean 6.79 1.74 22 
Total 40.77 10.42 157 
Data Source: Ghana Revenue Authority (GRA); United Nations Comtrade Database (accessed: December 2020) 
Notes: Bauxite (HS: 2606.00.0000). Interquartile range is calculated for unit prices (USD per kg) using -level GRA 
data. 
Negative numbers indicate over-invoicing while positive numbers indicate under-invoicing for the PCTG 
methodology. 
 
Estimates for undervalued exports: The estimates indicate that the amount of undervalued 
bauxite is USD 40.77 million (or 18.9% of the total export value) between 2011 and 2017. 
Although endogenously determined, the interquartile range filter provides better estimates of 
trade mis-invoicing where the lack of benchmark prices makes it not feasible to estimate the arm’s 
length price range using commodity market prices. The undervaluation could be attributed to 
the risk of transfer pricing from the multi-national company operating in the sector.   
115 
 
University of Ghana http://ugspace.ug.edu.gh 
Estimates for overvalued exports: Overvaluation of bauxite is approximately USD 10.42 
million, constituting roughly 4.7% of the total value of bauxite exported.  
Comparison with partner country trade gap estimates: In addition, partner-country trade gap 
estimations are done for bauxite exports from Ghana and the results are reported in Table 4:4 
Seventy-four percent of Ghana's total exports are undervalued according to the PCTG method 
(approximately USD 157 million). This may be the result of trade gaps and inaccurate reporting, 
as the export and import data from the UN COMTRADE database appears to list different 
countries. The Netherlands and the United Kingdom, for instance, appear on the list of countries 
that import bauxite from Ghana but not on the list of countries to which Ghana exports bauxite, 
whereas Angola and Gabon appear on the list of countries to which Ghana exports bauxite but 
not on the list of countries that import bauxite from Ghana. In addition, there may be trade gaps 
because some major trading partners, such as China, are missing years from the statistics. In 
contrast, 18 percent of total reported exports are overvalued according to the PCTG method 
(approximately USD 22 million). 
 
 
 
 
 
 
 
 
 
 
 
116 
 
University of Ghana http://ugspace.ug.edu.gh 
4.3.5 Estimates of Trade Mis-invoicing in Manganese (HS: 2602.00.0000) 
The annual and total undervaluation and overvaluation estimates of manganese are presented in 
Table 4:5. 
 
Table 4.5: Undervalued and Overvalued Exports – Manganese (HS: 2602.00.0000) 
Year Interquartile Range Interquartile Range Partner Country 
Filter: Below 25th Filter: Above 75th Trade Gaps (USD, 
pctle (USD, Million) pctle (USD) Million) 
 
2011 0.13 0.15 267 
 
2012 2.14 0.22 139 
 
2013 0.79 0.32 153 
 
2014 0.45 0.10 117 
 
2015 0.18 0.14 50 
 
2016 0.10 18.28 80 
 
2017 1.38 0.13 136 
 
Mean 0.74 2.76 135 
 
Total 5.17 19.34 942 
Data Source: Ghana Revenue Authority (GRA); United Nations Comtrade Database (accessed: December 2020) 
Notes: Manganese ores and concentrates (HS: 2602.00.0000).  Interquartile range is calculated for unit prices (USD 
per kg) using transaction-level data from GRA. 
Negative numbers indicate over-invoicing while positive numbers indicate under-invoicing for the PCTG 
methodology. 
 
 
 
117 
 
University of Ghana http://ugspace.ug.edu.gh 
Estimates of undervalued exports: The estimates indicate that the amount of undervalued 
manganese is USD 5.17 million (or 0.65% of the total export value) between 2011 and 2017. 
Given the nature of manganese trade, which is mainly bilateral, it is not possible to use 
benchmark prices to estimate trade mis-invoicing. The interquartile range price filter is used to 
generate the estimates. Although endogenously determined, the interquartile range filter 
provides better estimates of trade mis-invoicing when it is not possible to estimate the arm’s 
length price range using commodity market prices.  
 
Estimates of overvalued exports: Overvaluation of manganese is approximately USD 19.34 
million, constituting roughly 2.4% of the total value of manganese exported.  
 
Comparison with partner country trade gap estimates: In addition, partner-country trade gaps 
estimations are done for manganese exports from Ghana. The total trade gaps reported amount 
to USD 942 million. This is due to the huge gaps observed between the volume of exports and 
imports reported. For example, the amount of exports to China reported from Ghana in the UN 
COMTRADE data is only half of the total imports reported from China under imports from 
Ghana. Furthermore, some of the countries that Ghana exports manganese to, such as Australia, 
Romania, and the United Kingdom, are not on the list of countries that import manganese from 
Ghana, and some of the countries that import manganese from Ghana, such as Bulgaria, Slovakia, 
and Vietnam, are not on the list of countries that Ghana exports manganese to. It is possible that 
the huge asymmetries reported could be as a result of data inconsistencies. 
  
 
118 
 
University of Ghana http://ugspace.ug.edu.gh 
4.4 DETERMINANTS OF TRADE MIS-INVOICING 
This section provides some regression analysis of potential drivers of abnormal pricing in exports 
of gold from Ghana. The analysis is in the form of descriptive statistics and regression analysis. 
The regression analysis is done using Ordinary Least Squares (OLS) and Random Effects. Quantile 
regressions are also conducted as a robustness check.  
 
4.4.1 Descriptive Statistics 
Table 4.6 contains the descriptive statistics of the variables used for the regression analysis. Due 
to the high values of the variables showing overvaluation, undervaluation and current account 
balance, the log values are reported in this table. A table showing the actual values is included in 
Appendix 2. 
Table 4.6 Descriptive Statistics  
 Variable  Obs  Mean  Std. Dev.  Min  Max 
Undervaluation (Log) 463 11.536 4.457 -5.297 20.554 
Overvaluation (Log) 356 11.116 3.2 -5.297 16.566 
Current Account Bal (Log) 373 24.012 1.813 17.736 26.466 
Corruption 784 58.622 20.02 23 92 
Inflation 773 3.443 5.148 -1.667 58.933 
Tax Differential 560 0.019 0.081 -0.08 0.3 
Interest Rate 496 3.999 7.578 -12.857 41.76 
GDP Growth 784 2.713 2.742 -10.945 15.816 
 Transfer Pricing Risk 596 3.032 0.802 0.787 4.049 
Transfer Pricing Rank 596 0.638 0.291 0.017 1 
 
119 
 
University of Ghana http://ugspace.ug.edu.gh 
4.4.2 Correlation Analysis 
The correlation table below shows that the correlation between the various variables is less than 
0.5. The correlation between Transfer Pricing Risk and Transfer Pricing Rank is 0.961 so the two 
variables are used in two different regressions because of the high correlation between them.  
Table 4.7 Correlation Table 
Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) 
(1)Undervaluation 1.000          
(2) Overvaluation 0.084 1.000         
(3) Current Account  0.143 0.087 1.000        
(4) Corruption -0.082 0.096 0.114 1.000       
(5)Inflation 0.107 -0.064 -0.110 -0.480 1.000      
(6) Tax Differential 0.074 -0.015 0.132 0.181 -0.133 1.000     
(7) Interest Rate 0.019 -0.030 0.076 -0.151 -0.081 0.269 1.000    
(8) GDP Growth -0.017 -0.061 0.096 -0.113 -0.168 0.017 -0.109 1.000   
(9) Transfer Pricing Risk -0.008 0.071 0.048 -0.377 0.181 0.418 0.025 0.084 1.000  
(10) Transfer Pricing Rank 0.015 0.099 0.150 -0.344 0.148 0.386 0.068 0.104 0.961 1.000 
 
 
 
 
 
 
 
 
120 
 
University of Ghana http://ugspace.ug.edu.gh 
4.4.3 Regression Analysis 
The regression analysis uses the identified variables to estimate the drivers of trade mis-invoicing 
in commodity exports from Ghana. The regression is an unbalanced panel data regression for 2011 
to 2017.  The estimated regression models are Ordinary Least Squares (OLS) and Random Effects 
(RE). The Hausman (1978) test was conducted to determine the appropriate regression to be 
chosen between the Random Effects Model and the Fixed Effects Model. The test statistic for all 
the regressions were not significant at 1%, implying that the Random Effects model is preferred 
over the Fixed Effects model. The results of the Hausman test are shown in the Random Effects 
regression table and in Appendix 1.  
 
Table 4.8 Ordinary Least Squares (OLS) Regression   
 Overvaluation Overvaluation Undervaluation Undervaluation 
VARIABLES with TPRisk With TPRank  with TPRisk with TPRank 
     
Current Account Balance 1.300*** 1.126** 0.686** 0.735** 
 (0.470) (0.430) (0.333) (0.323) 
Corruption 0.0217 0.000644 -0.000160 0.0124 
 (0.0412) (0.0344) (0.0309) (0.0274) 
Inflation -0.841*** -0.852*** 0.0305 -0.000583 
 (0.226) (0.212) (0.172) (0.170) 
Tax Differential -44.07*** -44.90*** 0.546 -5.088 
 (13.76) (12.21) (11.15) (10.44) 
Interest Rate 0.302 0.250 -0.163 -0.104 
 (0.303) (0.274) (0.215) (0.207) 
GDP Growth -0.782*** -0.819*** -0.530*** -0.581*** 
 (0.220) (0.206) (0.174) (0.173) 
Transfer Pricing Risk 3.470***  -0.283  
 (0.950)  (0.731)  
REER -0.0606 -0.0636 0.0784** 0.0801** 
 (0.0457) (0.0433) (0.0350) (0.0349) 
Transfer Pricing Rank  11.32***  0.970 
  (2.369)  (1.924) 
Constant -22.18* -12.91 -10.89 -14.21* 
 (12.59) (10.19) (9.039) (7.770) 
     
Observations 79 79 85 85 
Prob>F 0.0009 0.0000 0.0018 0.0017 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
121 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 4.9 Random Effects Regression 
 Overvaluation Overvaluation Undervaluation Undervaluation 
with TPRisk with TPRank with TPRisk with TPRank 
VARIABLES     
     
Current Account 1.300*** 1.126*** 0.686** 0.735** 
Balance 
 (0.470) (0.430) (0.333) (0.323) 
Corruption 0.0217 0.000644 -0.000160 0.0124 
 (0.0412) (0.0344) (0.0309) (0.0274) 
Inflation -0.841*** -0.852*** 0.0305 -0.000583 
 (0.226) (0.212) (0.172) (0.170) 
Tax Differential -44.07*** -44.90*** 0.546 -5.088 
 (13.76) (12.21) (11.15) (10.44) 
Interest Rate 0.302 0.250 -0.163 -0.104 
 (0.303) (0.274) (0.215) (0.207) 
GDP Growth -0.782*** -0.819*** -0.530*** -0.581*** 
 (0.220) (0.206) (0.174) (0.173) 
Transfer Pricing Risk 3.470***  -0.283  
 (0.950)  (0.731)  
REER -0.0606 -0.0636 0.0784** 0.0801** 
 (0.0457) (0.0433) (0.0350) (0.0349) 
Transfer Pricing  11.32***  0.970 
Rank 
  (2.369)  (1.924) 
Constant -22.18* -12.91 -10.89 -14.21* 
 (12.59) (10.19) (9.039) (7.770) 
     
Observations 79 79 85 85 
Prob>F 0.0002 0.0000 0.0005 0.0005 
Hausman test, χ2 5.031 6.765                6.5 8.184 
P-value 0.754 0.562 0.591 0.416 
 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
 
Tables 4.8 and 4.9 document the regression results from the Ordinary Least Squares (OLS) and 
the Random Effects regression, respectively. The results confirm each other. 
 
122 
 
University of Ghana http://ugspace.ug.edu.gh 
Due to the simultaneous nature of the relationship between the two variables, however, these 
results should be interpreted with caution. 
 
Regarding overvaluation, it is observed that the Current Account Balance exhibits a positive 
correlation with both Transfer Pricing Risk and Transfer Pricing Rank in relation to overvaluation. 
This relationship is statistically significant. Our results are contrary to Patnaik et al 2012 who 
expect an inverse relationship between the Current Account and trade mis-invoicing. However, we 
hypothesise that the Current Account Balance could incentivise countries to over-value exports 
from Ghana.  
 
However, Inflation, the Tax Differential, GDP Growth, Transfer Pricing Risk and Transfer Pricing 
Rank all exhibit a negative but statistically significant relationship with overvaluation. The 
constant also shows a negative and significant relationship with overvaluation when Transfer 
Pricing Risk is used. Marur (2019) hypothesizes that Higher GDP growth is associated with a 
lowering trade mis-invoicing which these results confirm. It is expected that improvements in the 
condition of the partner country will diminish the incentive to mis-invoice. Due to the simultaneous 
nature of the relationship between the two variables, however, these results should be interpreted 
with caution. The results for inflation are in line with Kellenberg and Levinson 2018, who expect 
inflation to have a negative relationship with trade mis-invoicing because of firms’ having 
incentives to use trade mis-invoicing to move capital.  The results of the Tax Differential deviate 
from Marur (2019), who find that the tax differential has a positive relationship with trade mis-
invoicing. However, that study also finds that in some circumstances the tax differential rate has a 
negative relationship with trade mis-invoicing as this study finds. 
123 
 
University of Ghana http://ugspace.ug.edu.gh 
For undervaluation, it was observed that Current Account Balance and the REER have a positively 
significant relationship with undervaluation while GDP growth has a negative but statistically 
significant relationship with undervaluation. The constant shows a negative and significant 
relationship with undervaluation when Transfer Pricing Rank is used. Kwaramba et al. (2016) 
found that a high value of exchange rate signals a high level of illicit financial flows which 
confirms the results that the REER has a positive relationship with undervaluation. Similarly, as 
earlier indicated, a positive relationship is expected between the Current Account Balance and 
trade mis-invoicing. Also, Marur (2019) indicates a positive relationship between trade mis-
invoicing and GDP growth.  
 
4.4.2 Robustness Checks: Quantile Regression  
The study goes further to estimate a quantile regression as a robustness check for the earlier 
regressions done. Quantile regression is a highly versatile statistical modeling approach because it 
uses a general linear model to fit conditional quantiles of the response without assuming a 
parametric distribution.  
 
The purpose of this thesis was to examine the value chain of specific export commodities from 
Ghana in order to identify the risks for international financial flows (IFFs) in those commodities, 
measure trade mis-invoicing in Ghana's commodity exports using both the Partner Country Trade 
Gap Method and the Price Filter method and estimate the determinants of trade mis-invoicing in 
these commodities. This was performed at the commodity level, which is uncommon for studies 
of this type. 
 
124 
 
University of Ghana http://ugspace.ug.edu.gh 
Quantile regression is an alternative regression technique used when the key assumptions of linear 
regression are not met, there are outliers in the data, the residuals are not normal, there is an 
increase in error variance with increase in outcome variable, or to estimate the median or any 
quantile (Koenker & Hallock, 2001; Yu et al., 2003). In general, quantile regression results 
confirmed the results obtained from the OLS and Random Effects regressions although there are 
some deviations. 
 
Within the quantile regression for overvaluation with Transfer Pricing Risk, a negative relationship 
between overvaluation and inflation are observed within the 50th and 75th quantiles. Although a 
negative relationship was observed with inflation in the OLS and Random Effects regressions, the 
quantile regression clarifies that the relationship only exists within specific quantiles. Similarly, 
the tax differential exhibits a negative relationship with overvaluation but within the 50th and 90th 
quantiles only while GDP growth exhibits a negative relationship with overvaluation in the 50th 
quantile.  However, Transfer Pricing Risk exhibits a positive relationship with overvaluation 
within the 20th, 50th and 75th quantiles.  
 
 
 
 
 
 
 
 
 
 
125 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 4.10 Quantile Regression for Overvaluation using Transfer Pricing Risk 
(0.25 (0.75 
VARIABLES (0.5 Quantile) (0.9 Quantile) 
Quantile) Quantile) 
     
     
Current Account Balance 0.343 0.760 0.456 -0.515 
 (0.924) (0.841) (0.669) (0.688) 
Corruption 0.0530 0.0140 0.0236 0.0149 
 (0.0566) (0.0458) (0.0475) (0.0665) 
Inflation -0.441 -0.699** -0.564** -0.342 
 (0.320) (0.300) (0.280) (0.211) 
Tax Differential -32.85 -36.94** -19.07 -18.69* 
 (21.43) (15.49) (15.29) (10.86) 
Interest Rate 0.214 0.0827 0.183 -0.236 
 (0.322) (0.467) (0.484) (0.641) 
GDP Growth -0.463 -0.564** -0.399 -0.00306 
 (0.301) (0.261) (0.248) (0.157) 
Transfer Pricing Risk 3.085* 2.620* 1.843* 0.808 
 (1.753) (1.390) (1.014) (0.966) 
REER -0.0280 -0.0386 -0.0311 -0.0168 
 (0.0813) (0.0712) (0.0610) (0.0571) 
Constant -5.746 -8.195 0.628 27.29 
 (21.43) (20.31) (15.10) (16.41) 
     
Observations 79 79 79 79 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
 
 
In the quantile regression for overvaluation using Transfer Pricing Rank, the Current Account 
Balance shows a positive relationship in the 50th quantile while Transfer Pricing Rank exhibits a 
positive relationship with overvaluation in the 25th, 50th, and 75th quantiles. Inflation has a negative 
relationship with overvaluation at the 25th quantile, while the tax differential exhibits a negative 
relationship in all quantiles.  GDP growth shows a negative relationship at the 25th and 50th 
quantiles while the exchange rate shows a negative relationship at the 25th quantile. 
126 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 4.11 Quantile Regression for Overvaluation using Transfer Pricing Rank 
VARIABLES (0.25 Quantile) (0.5 Quantile) (0.75 Quantile) (0.9 Quantile) 
     
     
Current Account Balance 0.635 0.884* 0.459 -0.409 
 (0.498) (0.527) (0.533) (0.745) 
Corruption 0.0729** 0.0173 0.00690 0.0127 
 (0.0340) (0.0325) (0.0628) (0.0933) 
Inflation -0.519*** -0.664*** -0.539* -0.573 
 (0.134) (0.179) (0.293) (0.412) 
Tax Differential -44.36*** -33.92*** -29.71** -22.28* 
 (10.41) (11.91) (14.70) (11.90) 
Interest Rate 0.387 0.213 0.0719 0.163 
 (0.245) (0.301) (0.459) (0.630) 
GDP Growth -0.646*** -0.595*** -0.437 -0.261 
 (0.180) (0.193) (0.330) (0.476) 
Transfer Pricing Rank 13.56*** 9.301*** 6.740*** 4.593 
 (2.346) (1.601) (2.149) (3.950) 
REER -0.0509* -0.0352 -0.0260 -0.0186 
 (0.0264) (0.0354) (0.0565) (0.0480) 
Constant -10.74 -10.44 2.305 24.66 
 (13.95) (11.78) (13.66) (23.98) 
     
Observations 79 79 79 79 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
 
 
 
Undervaluation was observed that under the current account balance has a positive relationship 
with undervaluation in the 25th quantile while GDP growth has a negative relationship with 
undervaluation in the 25th and 50th quantiles. The exchange rate has a negative relationship with 
undervaluation in the 25th quantile but a positive relationship in the 50th quantile.  
 
 
 
 
127 
 
University of Ghana http://ugspace.ug.edu.gh 
Table 4.12 Quantile Regression for Undervaluation using Transfer Pricing Risk 
VARIABLES (0.25 Quantile) (0.5 Quantile) (0.75 Quantile) (0.9 Quantile) 
     
     
Current Account Balance 0.695* 0.497 0.273 0.350 
 (0.349) (0.322) (0.662) (1.020) 
Corruption 0.0186 -0.0316 0.0602 -0.105 
 (0.0295) (0.0307) (0.0772) (0.101) 
Inflation 0.0538 0.0313 0.463 0.337 
 (0.132) (0.168) (0.338) (0.357) 
Tax Differential 0.923 2.346 4.143 -21.62 
 (10.46) (12.20) (22.21) (22.65) 
Interest Rate -0.103 -0.255 0.0668 -1.109 
 (0.308) (0.285) (0.462) (0.746) 
GDP Growth -0.584*** -0.633*** -0.192 -0.155 
 (0.147) (0.199) (0.274) (0.344) 
Transfer Pricing Risk 0.230 -0.448 -0.951 -2.440 
 (0.653) (0.560) (1.744) (1.965) 
REER 0.0870*** 0.119*** 0.0738 -0.0316 
 (0.0306) (0.0355) (0.0463) (0.0642) 
Constant -16.23 -7.555 -2.816 25.14 
 (10.27) (8.233) (20.87) (29.97) 
     
Observations 85 85 85 85 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
 
 
For undervaluation using Transfer Pricing Rank, it was observed that the current account balance 
has a positive relationship with undervaluation at the 50th quantile while the exchange rate has a 
positive relationship in the 25th and 50th quantiles. GDP growth has a negative relationship in the 
25th  and 50th quantiles while the constant shows a negative relationship in the 50th quantile. 
 
 
 
 
 
128 
 
University of Ghana http://ugspace.ug.edu.gh 
 Table 4.13 Quantile Regression for Undervaluation using Transfer Pricing Rank 
VARIABLES (0.25 Quantile) (0.5 Quantile) (0.75 Quantile) (0.9 Quantile) 
     
     
Current Account Balance 0.380 0.681** 0.418 0.633 
 (0.296) (0.299) (0.686) (0.831) 
Corruption 0.0158 -0.0175 0.0618 0.0628 
 (0.0400) (0.0441) (0.0602) (0.0941) 
Inflation -0.182 -0.0765 0.173 0.239 
 (0.194) (0.196) (0.286) (0.389) 
Tax Differential -0.443 -2.929 0.703 -15.99 
 (15.84) (14.31) (17.58) (27.88) 
Interest Rate 0.0539 -0.0747 0.235 -0.103 
 (0.349) (0.300) (0.405) (0.635) 
GDP Growth -0.744*** -0.642** -0.417 -0.199 
 (0.273) (0.268) (0.297) (0.360) 
Transfer Pricing Risk 2.425 0.371 -0.708 0.882 
 (1.543) (1.942) (3.593) (6.087) 
REER 0.104*** 0.106*** 0.0893* -0.00423 
 (0.0331) (0.0342) (0.0459) (0.0573) 
Constant -10.15 -13.46* -9.576 -4.158 
 (7.662) (7.375) (15.14) (22.12) 
     
Observations 85 85 85 85 
Standard errors in parentheses 
*** p<0.01, ** p<0.05, * p<0.1 
 
 
 
4.5 CHAPTER SUMMARY 
This section presented the value chain analysis and identification of risks for IFFs in the selected 
commodities. The chapter goes further to present and analyse the estimates of IFFs using both the 
Partner Country Trade Gap Method and the Price Filter Method. In addition, the section contains 
a regression analysis of the drivers of trade mis-invoicing in the export of selected commodities 
from Ghana.   
 
 
 
129 
 
University of Ghana http://ugspace.ug.edu.gh 
CHAPTER FIVE 
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 
5.1 INTRODUCTION  
This chapter presents the summary, conclusions, and recommendations of the thesis. It also 
outlines the contributions of the thesis to knowledge.  
5.2 SUMMARY 
The purpose of this thesis was to examine the value chain of specific export commodities from 
Ghana in order to identify the risks for international financial flows (IFFs) in those commodities, 
measure trade mis-invoicing in Ghana's commodity exports using both the Partner Country Trade 
Gap Method and the Price Filter method and estimate the determinants of trade mis-invoicing in 
these commodities. This was performed at the commodity level, which is uncommon for studies 
of this type. The commodities chosen were gold, cocoa, bauxite, and manganese. The estimates 
show high trade mis-invoicing, especially for gold, cocoa paste and bauxite. This is in line with 
the foreign ownership of these firms which tends to create incentives for multi-national transfer 
pricing abuses and other practices which may go contrary to the use of the arm’s length principle.  
 
The study analyses gold doré exports from Ghana, which incorporate a combination of gold, 
silver and impurities, including lead and arsenic, whose composition can vary depending on the 
source mine. The transaction-level export data are matched with mine-level information on the 
gold and silver content in their doré production.  
 
130 
 
University of Ghana http://ugspace.ug.edu.gh 
The London Bullion Market Association (LBMA) daily spot prices for 99.9% pure, refined gold 
and silver are identified as the relevant benchmark price for Ghanaian exports. The mine-level 
purity information is combined with the commodity exchange prices for gold and silver to estimate 
the relevant benchmark price for gold doré exports from Ghana. Also, commodity experts estimate 
the variance in contractual pricing terms under normal business conditions to estimate the arm’s 
length price range. The results indicate that Ghanaian exports are undervalued by approximately 
11% of the total value of gold exported (USD 35.6 billion).   
Due to the lack of accurate benchmark prices and comparability criteria, the interquartile 
range price filter is used to estimate the arm's length price range for exports of cocoa paste, 
bauxite and manganese. 
 
The presence of international mining and trading firms in Ghana's liberalised gold sector can 
explain this overall finding of significant undervaluation. Typically, these firms are 
domiciled in jurisdictions where they benefit from lower effective corporate tax rates relative to 
Ghana, giving them an incentive to under-report their taxable income. For example one of the 
largest gold mining and exporting firm operating in Ghana is domiciled in Colorado, the United 
States of America, where the corporate tax rate equals 4.5% compared to 25% in Ghana.  
 
Furthermore, the semi-processed gold doré production from Ghana is first exported to 
destinations with gold refineries which also usually offer tax incentives, including reduced tax 
rates and complete tax exemptions for precious metal imports. For example gold doré imports are 
exempt from Value Added Tax in Switzerland which is the world’s largest gold refining country. 
131 
 
University of Ghana http://ugspace.ug.edu.gh 
The results indicate that the total estimated tax base erosion from Ghana due to the undervaluation 
of gold exports is USD 957.3 million. 
 
Next, the case of cocoa beans exports from Ghana that predominantly comprises superior quality 
fermented beans are analysed. Based on interviews with Cocoa Marketing Company (CMC) 
experts, the benchmark price for these exports is identified as the London International Financial 
Futures and Options Exchange (LIFFE) prices. Then information on the product-specific price 
premiums due to various certifications and transportation costs, which are normally included in 
reported Customs valuation, as well as expected variance in pricing terms under normal business 
conditions are included to estimate the arm’s length price range. The results indicate that Ghanaian 
cocoa beans exports are undervalued by USD 126.6 million which represents 1.0% of the total 
export value of USD 12.6 billion between 2011 and 2017. The low magnitude of abnormal pricing 
in cocoa beans exports can be explained by the Government of Ghana’s monopoly over cocoa 
bean exports from the country. Due to an absence of private actors in this partly liberalised cocoa 
sector of Ghana, the wholly state-owned Cocoa Marketing Company directly markets this 
commodity in the international markets and does not face economic incentives for trade 
mispricing. 
 
Due to the lack of accurate benchmark prices and comparability criteria, the interquartile range 
price filter is used to estimate the arm's length price range for exports of cocoa paste, bauxite, 
and manganese. 
The results indicate that 7.2% of the total export of cocoa paste (USD 1.8 billion) was 
undervalued i.e. found below the 25th percentile of the per unit price distribution of the product. 
132 
 
University of Ghana http://ugspace.ug.edu.gh 
Despite the limited nature of this methodology, the relatively high undervaluation of cocoa paste 
exports represents a significant risk for illicit outflows via transfer mispricing. Unlike the 
Ghanaian state monopoly over cocoa bean exports, private traders and multinational firms are 
allowed to manufacture and export cocoa products, including cocoa paste, cocoa powder and cocoa 
butter. These exporting firms face economic incentives for tax-motivated profit shifting by 
engaging in export undervaluation of their international sales, usually to affiliated marketing 
firms operating under a common ownership structure.  
 
The estimates indicate that the amount of undervalued bauxite is USD 40.77 million (or 18.9% 
of the total export value) between 2011 and 2017, while the amount of undervalued manganese 
is USD 5.17 million (or 0.65% of the total export value) between 2011 and 2017. Overvaluation 
of manganese is approximately USD million, constituting roughly 2.4% of the total value of 
manganese exported. 
 
Also, the price filter estimates are compared with the more commonly used partner-country trade 
gaps based on asymmetries between Ghana’s exports and the imports reported by its trading 
partners in the rest of the world.  
Using annual product-level data from Ghana Customs and the United Nations COMTRADE 
database, the study identifies significant asymmetries between reported Ghanaian exports and 
imports of gold, cocoa beans, cocoa paste, bauxite and manganese. 
 
Using annual product-level data from Ghana Customs and the United Nations COMTRADE 
database, the study identifies significant asymmetries between reported Ghanaian exports and 
133 
 
University of Ghana http://ugspace.ug.edu.gh 
imports of gold, cocoa beans, cocoa paste, bauxite, and manganese. In the case of gold, the trade 
gaps indicate that reported Ghanaian exports exceed what the rest of the world reports as imports 
from Ghana between 2011 to 2017, i.e. export over-valuation. This is contrary to our price filter 
estimates for the same period which indicate significant undervaluation in gold exports. However, 
an investigation of the aggregate trade data highlights that the observed trade gaps are primarily 
driven by missing imports data in the UN COMTRADE database from significant destinations 
for Ghanaian gold, including South Africa, Switzerland and the United Arab Emirates. 
 
In the case of cocoa beans, the estimated trade gaps indicate that reported Ghanaian exports are 
significantly lower than the reported imports by the rest of the world from Ghana. The estimated 
magnitude of undervaluation is significantly higher than what is estimated using transaction-level 
export data. The hypothesis is that the that drives these differences in international trade statistics, 
whereby the exporting countries record the destination as the intermediate shipping hub from 
where commodities are then re-exported to the final destination. This is a significant feature of 
the international trade in cocoa beans, which can be purchased and stored for multiple years before 
being marketed and shipped to the final destination.  
 
Then it is observed that significant data misreporting in product-level trade statistics for cocoa 
paste by Ghana Customs does not allow for estimating trade gaps. Ghana Customs records all 
export transactions for cocoa paste, cocoa butter and cocoa powder under the tariff category of 
cocoa paste, leading to incomparability with the import statistics of its trading partners. Overall, 
this comparison highlights the strong limitations of applying this methodology in the case of 
developing countries with poor statistical capacities and misreported trade data. 
134 
 
University of Ghana http://ugspace.ug.edu.gh 
 
In addition, a regression analysis is used to examine the determinants of trade mis-invoicing using 
variables such as current account balance, transfer pricing risk, transfer pricing rank, inflation and 
the tax differential.  
It is observed that contrary to the literature and there is a positive relationship with the current 
account balance and over-valuation.  Similarly, in a deviation from the literature, there is a negative 
relationship between over-valuation and the tax differential. However, as expected, a negative 
relationship is observed between GDP Growth, Transfer Pricing Risk, Transfer Pricing Rank and 
Inflation. These imply that improvements in the macro-economic conditions of countries tend to 
reduce the incentive for over-valuation.  
 
For undervaluation, it was observed that under the current account balance and the REER have a 
positive relationship with undervaluation while GDP growth has a negative relationship with 
undervaluation. The findings for the current account balance and the exchange rate confirm the 
literature, while the finding on GDP growth deviates from the literature.   
 
A quantile regression was also run as a robustness check. Overall, the quantile regression results 
confirm the results obtained from the OLS and Random Effects regressions. 
 
 
 
 
135 
 
University of Ghana http://ugspace.ug.edu.gh 
5.3 CONCLUSION  
Ghana, as a resource-rich developing nation, expects to earn much revenue from its natural 
resources to fund socio-economic growth. However, trade mis-invoicing undermines the ability of 
the natural resource sector to contribute to growth.  
 
The first goal of this study was to analyse the commodity value chains for some of Ghana’s most 
significant export commodities: cocoa, gold, bauxite and manganese using a rigorous, 
interdisciplinary approach. The study identified that the operations of multinational companies 
represent a high risk for trade mis-invoicing. 
 
The next goal was to estimate trade mis-invoicing in the selected commodities using the Price 
Filter Method with transaction level data from the Ghana Revenue Authority and the Partner 
Country Trade Gap (PCTG) method using data from UN COMTRADE. Finally, the study 
analyzed the determinants of trade mis-invoicing using regression analysis. The results of this 
study confirm that trade mis-invoicing is an urgent concern for Ghana.  
 
During the study, it was observed that the data capture process for export commodities leads to 
several errors due to ignorance on the part of some actors involved in the data capture process. For 
instance, it was observed that the generic term Gold Bullion is used for different types of gold 
exported from Ghana. This is a potential source of revenue loss for the country.  
 
It was also observed that different institutions involved in the commodity value chains do not share 
information thereby creating situations where disparate data is collected for different reasons rather 
136 
 
University of Ghana http://ugspace.ug.edu.gh 
than having an integrated system that will ensure harmonisation. This situation could potentially 
lead to a loss of revenue for the state. 
 
5.4 CONTRIBUTIONS TO KNOWLEDGE 
This study makes some useful contributions to knowledge. This current study is the first to estimate 
trade mis-invoicing from Ghana using multiple commodity exports and using transaction level 
data from the Ghana Revenue Authority from 2011 to 2017.  
 
Additionally, the study makes use of the method of extensive commodity research to understand 
the commodities studied and to identify the risks for IFFs in those commodities. This extensive 
qualitative research serves as the means for identifying applicable benchmark price series and 
defining the arm’s length price range after accounting for the expected level of price deviations.  
In the few previous studies that used price filter analyses, arbitrary criteria are used to set arm’s 
length price range.  
 
Another novelty of this study is that it generates estimates of trade mis-invoicing using the 
generally accepted Partner Country Trade Gap (PCTG) method at a commodity level for exports 
of gold, bauxite, manganese and cocoa beans and cocoa paste from Ghana. These estimates are 
also compared with the estimates generated from the Price Filter Method and both are analysed.  
 
In addition, the study contributes new evidence based on administrative microdata to analyse the 
possible tax revenue losses due to trade mis-invoicing, which could be indicative of the magnitude 
of IFFs in Ghana. 
137 
 
University of Ghana http://ugspace.ug.edu.gh 
The study further contributes to the trade mis-invoicing literature by utilising the estimates of trade 
mis-invoicing generated to analyse the determinants of trade mis-invoicing. This is important 
because many of the previous studies on trade mis-invoicing made use of estimates from the Parter 
Country Trade Gap (PCTG) Method to analyse the determinants of trade mis-invoicing. Hence, 
the use of transaction level estimates to analyse the determinants of trade mis-invoicing is quite 
novel.  
Another important contribution of this study is in respect of the Sustainable Development Agenda 
with a focus on SDG 16.4 (curb IFFs) and 17.1 (improve domestic revenue mobilisation). These 
are very key considerations for developing countries that have been reported to be losing funds 
through IFFs.  
 
The study also makes a significant contribution to policy formulation. A better appreciation of the 
issues regarding IFFs is necessary for providing policy prescriptions for improving domestic 
revenue mobilisation in developing countries. This current empirical study provides a good basis 
for doing that. 
 
5.5 RECOMMENDATIONS 
One of the study's recommendations is the need for relevant training for officials of the Customs 
Division of the Ghana Revenue Authority and all other individuals and organisations engaged in 
the capture of export data in Ghana. This is because of the gaps and errors observed in the data. 
Export data is critical for our development as a nation, and thus the same kind of attention paid to 
import data should be paid to export data.  
 
138 
 
University of Ghana http://ugspace.ug.edu.gh 
In addition, there is a need for the Customs Division of the Ghana Revenue Authority to provide 
the UN COMTRADE database with the requisite data especially for the years where there is 
missing data.  This is to ensure that estimates of trade mis-invoicing generated are not a result of 
trade discrepancies but are measuring trade mis-invoicing or other forms of IFFs.  
 
Further policy recommendations include the need to enhance the data capabilities of the different 
institutions exporting those goods, increased cooperation between different institutions in those 
sectors in order to harmonise the data collected, and constant improvement in staff skills. 
Information and communication technology tools, in particular computers, related software and 
access to vital databases, must also be updated to match those of private sector actors in order to 
enable the monitoring of tax assessments and payments.  
 
Regulators may also consider applying the investigation methodology based on price exchanges 
and statistical benchmarks as a way of risk-based customs, tax and transfer price audits.  
 
In addition, policymakers will need to prioritise the development of institutional expertise to map, 
control and block the sources of the resulting tax base erosion from Ghana due to trade mis-
invoicing. 
 
Finally, Civil Society Organisations (CSOs) may also need to collaborate more with other state 
institutions to research further into the area. 
 
 
139 
 
University of Ghana http://ugspace.ug.edu.gh 
5.6 LIMITATIONS AND DIRECTIONS FOR FURTHER RESEARCH 
The thesis was limited to five of Ghana’s export commodities. Further studies could look at 
estimating trade mis-invoicing in more commodities such as petroleum.  In addition, although the 
thesis offers strong results and a wide range of implications for regulatory authorities, policy 
makers and companies, future studies could evaluate the selected techniques, methods and 
methodology used. Finally, future studies could review the effect of trade mis-invoicing on various 
aspects of the economy such as foreign direct investment or the financial sector. 
 
  
140 
 
University of Ghana http://ugspace.ug.edu.gh 
REFERENCES 
Abbott, P. (2013). Cocoa and Cotton Commodity Chains in West Africa: Policy and Institutional 
Roles for Smallholder Market Participation, In: Rebuilding West Africa’s Food 
Potential, A. Elbehri (Ed.), FAO/IFAD. 
Abor, J.Y. (2014). Tackling Illicit Financial Outflows from Ghana Transfer Pricing 
Study.Integrated Social Development Centre (ISODEC) 
Acheampong, M. (2019, April 6) Personal Interview 
African Economic Outlook (2010). OECD Development Centre, African Development Bank and 
UNECA. 
Agbeyegbe, T. D., Stotsky, J. G. and WoldeMariam, A. (2006). “Trade Liberalization, Exchange 
Rate Changes, and Tax Revenue in Sub-Saharan Africa” Journal of Asian Economics, 
Vol. 17, pp. 261-284 
Aidt, T. S. (2003). “Economic Analysis of Corruption: A Survey.” The Economic Journal 113 
(491): F632–F652.  
Albaum, G., Strandskov, J., Duerr, E. and Dowd, L. (1989). International Marketing and Export 
Management, Addison-Wesley, Wokingham 
Amankwah, R. K., Wiafe, J. and Bekoe, A. (1999). “Production and Quality Assessment of 
Manganese Nodules from Nsuta Low Grade Carbonate Ore”, Ghana Mining Journal, 
Vol. 5, pp. 70-74 
Amerkhail, V. (2002). ‘Arm’s Length or Formulary Apportionment? Sometimes the Best Choice 
is Both’ 9(13) Tax Management Transfer Pricing Report 25; 
Ampofo, K.O. (2018, April 1). COCOBOD Inaugurates Audit Committee. GhanaWeb. Retrieved 
from: https://www.ghanaweb.com/GhanaHomePage/business/COCOBOD-inaugurates-
audit- committee-639364 
Andersen, O. (1993). On the internationalization process of firms: a critical analysis, Journal of 
International Business Studies, Vol. 24 No. 2, pp. 209-31. 
Andvig, J. C., O.-H. Fjeldstad, I. Amundsen, T. Sissener, and T. Søreide. (2001). “Corruption. A 
Review of Contemporary Research.” Report R 2001:7, Chr. Michelsen Institute, 
Bergen, Norway.  
Antràs, Pol and Ricardo Caballero (2009). Trade and capital flows: A financial frictions 
perspective. Journal of Political Economy, 117(4), 701–744. 
Ashitey, E. (2012). “Cocoa Report Annual.” Global Agricultural Information Network, GAIN 
Report Number GH1202, USDA, Washington. 
http://scholar.google.com/scholar?hl=en&q=Ashitey%2C+E.+2012.+%E2%80%9CCo
141 
 
University of Ghana http://ugspace.ug.edu.gh 
coa+Report+Annual.%E2%80%9D+Global+Agricultural+Information+Network%2C+
GAIN+Report+Number+GH1202%2C+USDA%2C+Washington.  
Baker, R. (2014). Hiding in Plain Sight: Trade Misinvoicing and the Impact of Revenue Loss in 
Ghana, Kenya, Mozambique, Tanzania, and Uganda: 2002-2011. Global Financial 
Integrity. 
Balma, L., Bawumia, M., & Ncube, M. (2018). Managing Natural Revenue for Sustainable 
Development in Ghana: Simulations using a DSGE Model. 
Bardhan, P. (1997). “Corruption and Development: A Review of Issues.” Journal of Economic 
Literature 35 (3): 1320–46.  
Beja, E, Junvith P, Ragusett J (2005). Capital Flight from Thailand, 1980–2000. In Epstein G 
(ed), Capital Flight and Capital Controls in Developing Countries. Chapter 6, pp. 143–
172. Edward Elgar Publishing 
Bell, K. A (2009) ‘Can Formulary Approach Fit into Arm’s Length Confines? Formulary 
Advocate Dubs OECD Draft Chapters a Significant Step’ 18(14) Tax Management Inc., 
767. 
Bensassi, S., & Jarreau, J. (2019). Price Discrimination in Bribe Payments: Evidence from 
Informal Cross-border Trade in West Africa. World Development, 122, 462-480. 
Berger, H., & Nitsch, V. (2008). Gotcha! A Profile of Smuggling in International Trade. 
Berger, H., & Nitsch, V. (2012). ‘Gotcha’! A Profile of Smuggling in International Trade. In C. 
C. Storti & P. De Grauwe (Eds.), Illicit trade and the global economy (chap. 4). 
Cambridge, MA: MIT Press. 
Berger, H., & Nitsch, V. (2012). Bilateral Imbalances in Europe, [Darmstadt Discussion Papers 
in Economics, No. 214/2012]. Darmstadt, Darmstadt University of Technology. 
Bhagwati, J. (1964a). On the Underinvoicing of Imports. Bulletin of the Oxford University 
Institute of Economics      &      Statistics,      27(4),      389–397.       
https://doi.org/10.1111/j.1468-0084.1964.mp27004007.x 
Bhagwati, J., (1964b). On the Under-invoicing of Imports. Bulletin of the Oxford University 
Institute of Statistics 26, 389–397. 
Blankenburg, S., & Khan, M. (2012). Governance and Illicit Flows. Draining, 21. 
Boyce, J. K., & Ndikumana, L. (2001). Is Africa a Net Creditor? New estimates of capital flight 
from severely indebted sub-Saharan African countries, 1970-96. Journal of 
Development Studies, 38(2), 27-56. 
Bracking, S. (2003). The political economy of chronic poverty. Available at SSRN 1754446. 
Bruce, E. (2018, September 27). Ghana, Cote d’Ivoire to Announce Floor Price for Cocoa on Oct 
1.Graphic     Online.     Retrieved     from:     
142 
 
University of Ghana http://ugspace.ug.edu.gh 
https://www.graphic.com.gh/business/business- news/ghana-cote-d-ivoire-to-announce-
floor-price-for-cocoa-on-oct-1.html 
Buckley, P.J. (1982). Multinational Enterprises and Economic Analysis, Cambridge University 
Press, London. 
Buckley, P.J. (1988). The limits of explanation: testing the internalisation theory of the 
multinational, Journal of International Business Studies, Vol. 19, pp. 181-93 
Buckley, P.J. and Casson, M. (1976). The Future of the Multinational Enterprise, Holmes and 
Meier, London. 
Buckley, P.J. and Casson, M. (1985). The Economic Analysis of the Multinational Enterprise: 
Selected Papers, Macmillan, London. 
Buehn, A., & Eichler, S. (2006). Uncovering Smuggling: Worldwide Evidence for Four Types of 
Trade Misinivoicing. Journal of Dresden University of Technology, Germany. 
Bulir, M. A. (1998). The Price Incentive to Smuggle and the Cocoa Supply in Ghana, 1950-
1996. International Monetary Fund. 
Carbonnier, G. & Mehrotra, R. (2020). Abnormal Pricing in International Commodity Trade: 
Empirical Evidence from Switzerland. R4D-IFF Working Paper Series, R4D-IFF-
WP01-2019. 
Carbonnier, G., & Zweynert de Cadena, A. (2015). Commodity Trading and Illicit Financial 
Flows. International Development Policy| Revue Internationale de Politique de 
Développement. 
Chen, Huan. (2009). A Literature review on the relationship between foreign trade and economic 
growth. International Journal of Economics and Finance 1(1): 127-130. 
Choi, Y. S., & McGauran, R. (2018). Conceptual Basis of Illicit Financial Flows, Trade Mis-
invoicing and Trade Discrepancy. Illicit Financial Flows via Trade Mis-invoicing, 14  
Chow, P. C. (1987). Causality between export growth and industrial development: Empirial 
evidence from the NICs. Journal of development Economics, 26(1), 55-63..  
Christian Aid (2009). False Profits: Robbing the Poor to Keep the Rich Tax-free. 
CMC Personnel (2019, June 3) Personal Interview 
Cobham, A., & Jansky, P. (2017). Illicit financial flows: An overview. Intergovernmental Group 
of Experts on Financing for Development. 
Cocoa Marketing Company (2019, June 3) Personal Interview  
Cocobod Personnel. (2019 April 6) Stakeholder Meeting 
Collin, M. (2020). Illicit Financial Flows: Concepts, Measurement, and Evidence, The World 
Bank Research Observer, https://doi.org/10.1093/wbro/lkz007  
143 
 
University of Ghana http://ugspace.ug.edu.gh 
Cotoc, C., Nițu, M., Șcheau, M. and Cozma, A. (2022). Efficiency of Money Laundering 
Countermeasures: Case Studies from European Union Member States. [online] 
Available at: <https://www.mdpi.com/2227-9091/9/6/120/htm#B10-risks-09-00120> 
[Accessed 10 July 2022].  
Creswell, J. W. (2012). Educational Research: Planning, Conducting, and Evaluating 
Quantitative and Qualitative Research (4th ed.). Boston, MA: Pearson Education, Inc 
CSSM (2009). Mining Together: Large-Scale Mining meets Artisanal Mining, Communities and 
Small- Scale Mining, World Bank, Washington, D.C. 
Dalli, D. (1994). The exporting process: the evolution of small and medium sized firms toward 
internationalisation, in Axinn, C.N. (Ed.), Advances in International Marketing, JAI 
Press, Greenwich, CT, pp. 85-110 
Dand, R. (2011). “The International Cocoa Trade” Third Edition. Woodhead Publishing. 
David. (2013) Personal Interview 
Davoodi, H. R. and Grigorian, D. A. (2007). Tax Potential vs Tax Effort: A Cross-Country 
Analysis of Armenia’s Stubbornly Low Tax Collection. IMF Working Paper, 07/106, 
Washington: IMF 
de Boyrie, M. E., Nelson, J. A., & Pak, S. J. (2007). Capital Movement Through Trade 
Misinvoicing: the Case of Africa. Journal of Financial Crime, 14(4), 474-489. 
de Boyrie, M. E., Pak, S. J., & Zdanowicz, J. S. (2005). Estimating the Magnitude of Capital 
Flight due to Abnormal Pricing in International Trade: The Russia–USA case. In 
Accounting Forum (Vol. 3, No. 29, pp. 249-270). 
De Wulf, L. (1981). Incidence of budgetary outlays: where do we go from here? Public Finance, 
36(1), 55-76.  
Demetriades, Panicos, and Vassileva, R. (2020). Money laundering and Central Bank 
Governance in the European Union. Journal of International Economic Law 23: 509–
33. 
Devine, K., O’Clock, P. & Seaton, L. P. (2006). ‘Estimating Impact of Formula Apportionment 
on allocation of Worldwide Income and the Potential for Double Taxation’, 19 
Advances in International Accounting, 115  
Dunning, J.H. (1980). Toward an eclectic theory of international production: some empirical 
tests, Journal of International Business Studies, Vol. 11 No. 1, pp. 9-31  
Eden, L. (2012). Transfer price manipulation. Draining development, 205-234. 
 
144 
 
University of Ghana http://ugspace.ug.edu.gh 
Elliott, J. and Emmanuel, C. (2002). International transfer pricing. In: Lymer, A. and Hasseldine, 
J. (eds.) International Taxation System. Kluwer Academic: Boston, USA, pp. 57-180. 
ISBN 9781402071577. 
Engel, C. D. (1998). ‘Would Formula Apportionment for International Transactions be More 
Efficient Approach’ (1998) 24(1) The International Tax Journal 23 
Ernst & Young, (2010). Transfer pricing global reference guide, London, U.K.: Ernst & Young. 
Etikan, I., Musa, S. A., & Alkassim, R. S. (2016). Comparison of Convenience Sampling and 
Purposive Sampling. American Journal of Theoretical and Applied Statistics, 5(1), 1-4. 
Farzanegan, M.R. (2009). Illegal Trade in the Iranian Economy: Evidence from a Structural 
Model. European Journal of Political Economy 25: 489-507 
Fayerweather, J. (1982). International Business Strategy and Administration, Ballinger, 
Cambridge, MA 
Feinschreiber, R. (2004). ‘Practical Aspects of Transfer Pricing’ in Robert Feinschreiber (ed) 
Transfer Pricing Methods: An Applications Guide  
Ferwerda, Joras, Saase, A. V., Unger, B., and Getzner, M. (2020). Estimating money laundering 
flows with a gravity model-based simulation. Scientific Report 10: 1–11. 
Fisman, R., Moustakerski, P., & Wei, S.-J. (2008). Outsourcing Tariff Evasion: A New 
Explanation for Entrepôt Trade. The Review of Economics and Statistics, 90(3), 587–
592. 
Fisman, R., Wei, S.-J., (2007). The Smuggling of Art, and the Art of Smuggling: Uncovering the 
Illicit Trade in Cultural Property and Antiques. NBER Working Paper No.W13446. 
National Bureau of Economic Research, Cambridge, MA. 
Forstater, M. (2015). Can Stopping “Tax Dodging” by Multinational Enterprises Close the Gap 
in Development Finance? CGD Policy Paper, 69, 32. 
Forstater, M. (2016). Illicit Flows and Trade Misinvoicing: Are we looking under the wrong 
lamppost? CMI Insight. 
Forstater, M. (2017a). Are the big numbers on illicit financial flows misleading, and does it 
matter? Lifting the Veil of Secrecy, 60. 
Forstater, M. (2017b). Gaps in Trade Data do not equal Criminal Money Laundering. Center for 
Global Development Website. http://www. cgdev. org/blog/gaps-trade-data-criminal-
money-laundering. Published. 
Forstater, M. (2018). Illicit Financial Flows, Trade Mis-invoicing, and Multinational Tax 
Avoidance: The Same or Different? CGD Policy Paper, 123, 29. 
Furusawa, T. and Yanagawa, N. (2013). International trade and capital movement under 
financial imperfection. mimeo 
145 
 
University of Ghana http://ugspace.ug.edu.gh 
Ghana Chamber of Mines (2018). 2017 Annual Report. Accra 
Ghana Chamber of Mines (2019). 2018 Annual Report. Accra 
Gilbert, C. L. (2009). Cocoa Market Liberalization in Retrospect. Review of Business and 
Economic Literature, 54(3), 294-313. 
Giles, J., & Williams, C. L. (2000). Export-led growth: a survey of the empirical literature and 
some non-causality results. Part 1. The Journal of International Trade & Economic 
Development, 9(3), 261-337. 
Global Financial Integrity. (2017). Illicit Financial Flows to and from Developing Countries: 
2005-2014. Retrieved from http://www.gfintegrity.org/report/illicit-financial-flows-to-
and-from-developing-countries-2005-2014/  
Goredema, C. (2011). Combating Illicit Financial Flows and Related Corruption in Africa: 
Towards a more Integrated and Effective Approach. U4 Issue, 2011(12).  
Gries, T., & Redlin, M. (2020). Trade and economic development: global causality and 
development-and openness-related heterogeneity. International Economics and 
Economic Policy, 17(4), 923-944. 
Gupta, S., Clement, B., Pivovarsky, A. and Tiongson, R. (2003). Foreign Aid and Revenue 
Response: Does the Composition of Aid Matters? IMF Working Paper 03/176, 
Washington: IMF 
Haar, J. and Ortiz-Buonafina, M. (1995). The internationalisation process and marketing 
activities: the case of Brazilian export firms, Journal of Business Research, Vol. 32, pp. 
175-81 
Hahm, J. and Shin K. (2009). Complementarity among international asset holdings. Journal of 
the Japanese and International Economies 23(1), 37–55. 
Hausman, J. (1978). Specification tests in econometrics. Econometrica, 46, 1251–1271. 
Hausermann, H., and Ferring, D. (2018). Unpacking Land Grabs: Subjects, Performances and the 
State in Ghana's ‘Small‐scale’ Gold Mining Sector. Development and Change, 49(4), 
1010-1033.  
Hayes, R. H., and Abernathy, W. J. (1980). Managing our way to economic decline. Harvard 
Bus. Rev. ;(United States), 58(4). 
Heckscher, E. (1919). The effect of foreign trade on the distribution of income. Ekonomisk 
Tidskrift, 21, 497–512. Reprinted as Chapter 13 in A.E.A. and Ellis, H.S. (1949). 
Readings in the theory of international trade, 272–300. Philadelphia: Blakiston. With a 
Translation in: H. Flam & Flanders M. J. (Eds.). (1991). Heckscher-Ohlin trade theory, 
43–69. Cambridge, MA: MIT Press.  
Hecksher, E. and Ohlin, B. (1933). Interregional and International Trade, Harvard University 
Press, Cambridge, MA 
146 
 
University of Ghana http://ugspace.ug.edu.gh 
Heggstad, K., & Fjeldstad, O. H. (2010). How Banks Assist Capital Flight from Africa: A 
Literature Review. CMI Report, 6. 
Hellerstein, J. R (1975). ‘The Unitary Business Principle and Multicorporate Enterprises- an 
Examination of the Major Controversies’ 27(4) Tax Executive 313. 
Helpman, E. and Razin A. (1978). Uncertainty and international trade in the presence of stock 
markets. Review of Economic Studies 45(2), 239–250 
Hill W. L. C. (2009). International Trade Theory PPT Chapter 5 pp 1-65; Global Business 
Today, The McGraw-Hill Companies, Inc 
Hirshleifer, J. (1956). “On the Economics of Transfer Pricing.” Journal of Business 29 (3): 172–
83. ———. 1957. “Economics of the Divisionalized Firm.” Journal of Business 30 (2): 
96–108. 
Hong, K. P., & Pak, S. J. (2017). Estimating Trade Mis-invoicing from Bilateral Trade Statistics: 
The Devil is in the Details. The International Trade Journal, 31(1), 3-28. 
Hong, K., H. Pak, C., & J. Pak, S. (2014). Measuring Abnormal Pricing–an Alternative 
Approach: The Case of US Banana Trade with Latin American and Caribbean 
Countries. Journal of Money Laundering Control, 17(2), 203-218. 
Hunter, M. (2020). "Illicit Financial Flows: Artisanal and Small-scale Gold Mining in Ghana and 
Liberia", OECD Development Co-operation Working Papers, No. 72, OECD 
Publishing, Paris, https://doi.org/10.1787/5f2e9dd9-en. 
Huszagh, S.M. and Huszagh, F.W. (1986). International barter and countertrade, International 
Marketing Review, Vol. 3 No. 2, Summer, pp. 16-28 
Hymer, S. (1970). United States investment abroad. Third Pacific Trade and Development 
Conference. 
Index Mundi, https://www.indexmundi.com/minerals/?product=bauxite (accessed on 
23/12/2019) 
ISSER (2018). State of the Ghanaian Economy Report in 2017. Institute of Statistical, Social and 
Economic Research, University of Ghana, Accra, Ghana.  
Jaara, O. O., Kadomi., A. M. (2017). Factors related to the Central Bnak instructions on money 
laundering. Journal of Money Laundering Control 20: 274–91 
Janský, P. (2013). Illicit Financial Flows and the 2013 Commitment to Development Index. 
Centre for Global Development (CGD) Policy Paper, 34.  
Jayantilal, Shital, Jorge S. F., Ferreira, A. (2017). A Portuguese anti-money laundering policy: A 
game theory approach. European Journal on Criminal Policy and Research 24: 559–74. 
Johannesen, N., & Pirttilä, J. (2016). Capital Flight and Development: An Overview of Concepts, 
Methods, and Data Sources (No. 2016/95). WIDER Working Paper. 
147 
 
University of Ghana http://ugspace.ug.edu.gh 
Jones, R. (1967). International capital movements and the theory of tariffs and trade. The 
Quarterly Journal of Economic,  81(1), 1–38. 
Kaplinsky, R., & Morris, M. (2003). A Handbook for Value Chain Research Brighton. UK: 
Institute for Development Studies. 
Kar, D., & Cartwright-Smith, D. (2010). Illicit Financial Flows from Africa: Hidden Resource 
for Development. 
Kar, D., & Freitas, S. (2012). Illicit Financial Flows from Developing Countries: 2001-2010. 
Washington, DC: Global Financial Integrity. 
Kar, D., & LeBlanc, B. (2013). Illicit Financial Flows from Developing Countries: 2002-2011. 
Washington, DC: Global Financial Integrity. 
Kar, D., and D. Cartwright-Smith. (2008). “Illicit Financial Flows from Developing Countries, 
2002–2006.” Global Financial Integrity, Washington, DC.  
Kellenberg, D. and Levinson, A. (2018). Misreporting trade: tariff evasion, corruption and 
auding standards. (No. w22593). National Bureau of Economic Research 
Kemp, M. (1966). The gain from international trade and investment: A neo-Heckscher-Ohlin 
approach. American Economic Review, 56(4), 788–809. 
Kesse, G. O. (1985). The Mineral and Rock Resources of Ghana, Balkema, Rotterdam, 610p. 
Khan, M. S., & Haque, N. U. (1985). Foreign Borrowing and Capital Flight: A Formal Analysis. 
Staff Papers, 32(4), 606-628. 
Khattry, B. and Rao, M. J. (2002). Fiscal Faux Pas? An Analysis of the Revenue Implications of 
Trade Liberalization. World Development Vol. 30, No. 8, pp. 1432-1444 
Khoury, S.J. (1984). Countertrade: forms, motives, pitfalls and negotiation requisites, Journal of 
Business Research, Vol. 12 No. 2, June, pp. 98-107 
Kind, H. J., K.-H. Midelfart-Knarvik, And G. Schjelderup (2005). Corporate tax systems, 
multinational enterprises and economic integration, Journal of International Economics 
65, 507–521. 
Koenker, R., and Hallock, K.F. (2001). "Quantile Regression." Journal of Economic 
Perspectives, 15 (4), 143-156. 
Kolavalli, S., & Vigneri, M. (2011). Cocoa in Ghana: Shaping the Success of an Economy. Yes, 
Africa can: success stories from a dynamic continent, 201-217. 
Kotil, E., and Konur F. (2010). The relationship between growth and foreign trade in Turkey: A 
Granger causality approach (1989-2007). Middle Eastern Finance and Economics 1, 
no.6: 32-40 
148 
 
University of Ghana http://ugspace.ug.edu.gh 
Kpodo, K. (2017, September 14). Update 2-Ghana to Maintain Cocoa Price, Raising Smuggling 
Risk. Reuters. Retrieved from: https://www.reuters.com/article/cocoa-ghana/update-2-
ghana- to-maintain-cocoa-price-raising-smuggling-risk-idUSL5N1LV4RM 
Kumar, V., (2012). Money Laundering: Concept, Significance and its Impact. [online] 
Core.ac.uk. Available at: <https://core.ac.uk/download/pdf/234624157.pdf> [Accessed 
11 July 2022].  
Kwaramba, M., Mahonye, N., & Mandishara, L. (2016). Capital Flight and Trade Misinvoicing 
in Zimbabwe. African Development Review, 28(S1), 50-64. 
Leontief, W. (1966). Input-output Economics, Oxford University Press, Oxford. 
Le,K., Nhien, A., Sammer, M., O‘Neil, A. B. M., Kechadi, T. (2016). An efficient search tool for 
anti-money laundering application of an multinational bank‘s dataset. arXiv 
arXiv:1609.02.31. 
Lerner, A. (1944). Economics of Control: Principle of Welfare Economics, Macmillan, New 
York, NY, USA. 
Lesclous R. (2002). “Électricité et aluminium en Afrique subsaharienne: stratégie des 
producteurs d'aluminium”, Outre-mers, 89(334-335), pp. 163-175 
Mahmood, Z. (1997). Research note: Determinants of Under-invoicing of Imports in Pakistan. 
Journal of International Development, 9(1), 85–96.  
Markam, M.  (2004). “Transfer pricing of intangible assets in the US, the OECD and Australia: 
Are profit-split methodologies the way forward?” University of Western Sydney Law 
Review 8,55. 
Markusen, J. R. (1983). Factor movements and commodity trade as complements. Journal of 
international economics, 14(3-4), 341-356. 
Marur, S. (2019). Drivers of Abnormal Pricing in Switzerland's Commodity-Trade. 
Mate-Kodjo, F. (2019, May 21) Personal Interview 
McLennan, P. G. (2012). Tackling Tax Evasion: Transfer Price Manipulation, Extractive Natural 
Resources and a Strategy for the Southern African Customs Union. 
McLure, C. E. (2002). ‘Replacing Separate Entity Accounting and the Arm’s Length Principle 
with Formulary Apportionment’56(12) Bulletin for International Fiscal Documentation 
586 
McMahon, G. and Schmidt-Hebbel, K. (2000). Macroeconomic Adjustment and Tax Reform in 
Developing Countries.” In Perry, G., Whalley, J. and McMahon, G. eds. Fiscal Reform 
and Structural Change in Developing Countries, Vol. 2, Macmillan Press Ltd, London. 
pp. 150-196. 
149 
 
University of Ghana http://ugspace.ug.edu.gh 
Mescall, D., & Klassen, K. J. (2018). How Does Transfer Pricing Risk Affect Premiums in 
Cross‐Border Mergers and Acquisitions? Contemporary Accounting Research, 35(2), 
830-865. 
Metal Focus Doré Flow Services Database, 2019, Metal Focus 
Milatovic, D., Gupta, R. C., Yin, Z., Zaja-Milatovic, S., & Aschner, M. (2011). Manganese. In 
Reproductive and Developmental Toxicology (pp. 439-450). Elsevier Inc. 
https://doi.org/10.1016/B978-0-12-382032-7.10034-7 
Minerals and Mining General Regulation 2012 (LI 2173).  
Minerals Commission (2017). 2016 Annual Report. Accra 
Minerals Commission (2018). 2017 Annual Report. Accra 
Mundell, R. (1957a). Transport Costs in International Trade Theory. Canadian Journal of 
Economics and Political Science, 23(3), 331-348. doi:10.2307/138957 
Mundell, R. (1957b). International trade and factor mobility. American Economic Review 47(3), 
321–335.  
Naheem, M. A. (2016). Money laundering: A primer for banking staff. International Journal of 
Disclosure and Governance 13: 135–56. 
Ndikumana, L. (2004). “Fiscal Policy, Conflict, and Reconstruction in Burundi and Rwanda”, In 
Addison, T. and A. Roe (eds.), Fiscal Policy for Development, Poverty, Reconstruction 
and Growth, Palgrave – MacMillan, pp. 274-302 
Ndikumana, L. (2016). Trade Misinvoicing in Primary Commodities in Developing Countries: 
The cases of Chile, Côte d’Ivoire, Nigeria, South Africa and Zambia. Geneva: 
UNCTAD. 
Ndikumana, L., Boyce, J. K., & Ndiaye, A. S. (2015). Capital flight from Africa: Measurement 
and drivers. In I. Ajayi and L. Ndikumana (Eds), Capital Flight from Africa: Causes, 
Effects and Policy Issues, 15–54. 
Neilsen, S. B; Raimondos-Moller, P. & Guttorm, S. (2001). Formulary Apportionment and 
Transfer Pricing under Oligopolistic Competition, Centre for Economic Studies, 
Working Paper No.491 
Nicolaou-Manias, K. (2016). Trade Mispricing for Five African Countries. GEG Africa Policy 
Brief-ing, October, Global Economic Governance Africa. http://www. gegafrica. 
org/publications/trademispricing-for-five-african-countries. 
Nitsch, V. (2016). Trillion Dollar Estimate: Illicit Financial Flows from Developing Countries 
(No. 227). Darmstadt Discussion Papers in Economics 
150 
 
University of Ghana http://ugspace.ug.edu.gh 
Nolintha, V., Sayavong, V., & Mehrotra, R. (2020). Abnormal Pricing in International 
Commodity Trade: Evidence from Lao P.D.R. R4D-IFF Working Paper Series, R4D-
IFF-WP02-2019. 
Nyarko, C. (2019, October 17) Personal interview 
Observatory of Economic Complexity (OEC) (2019).   https://oec.world/en/profile/country/gha/ 
Ocloo ,D. R. (2018). GRA loses GH¢4.5bn through Fraud at Ports. Graphic Online. 
https://www.graphic.com.gh/business/business-news/gra-loses-gh-4-5-bn-through-
fraud- at-ports.html 
OECD (Organisation for Economic Co-operation and Development). —. (2010). “Intra-firm 
Trade: A Work in Progress.” Document STD/TBS/WPTGS (2010)24 (September 8), 
Working Party on International Trade in Goods and Trade in Services Statistics, 
Statistics Directorate, OECD, Paris 
OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 1995. 
Ohlin, B. 1933. Interregional and international trade. Cambridge, MA: Harvard University Press 
Opare-Hammond, K. (2018, December 5) Personal Interview 
Oppong, N. Y. (2013). The Good and the Bad of Structural Adjustment Programme in Africa: 
Evidence from Ghanaian Gold Mining Industry. International Journal of Business and 
Behavioral Science, 3(10), 27-41. 
Organisation For Economic Co-Operation And Development (OECD). (2001). Higher Education 
Management. Journal Of The Programme On Institutional Management In Higher 
Education, 13 (2) 
Organisation for Economic Co-operation and Development.  (2017). OECD Transfer Pricing 
Guidelines for Multinational Enterprises and Tax Administrations 2017, OECD 
Publishing, Paris. http://dx.doi.org/10.1787/tpg-2017-en 
Organisation for Economic Co-operation and Development. (2014). Illicit Financial Flows from 
Developing Countries: Measuring OECD Responses. OECD Publishing. Retrieved 
from: 
https://search.oecd.org/corruption/Illicit_Financial_Flows_from_Developing_Countries
.pdf 
Owens, J (2005) "Should the Arm’s Length Principle Retire?’ International Transfer Pricing 
Journal 99; 
Oxford Business Group, (2013). Ghana: Showing its metals. Available at: 
http://www.oxfordbusinessgroup.com/news/ghana-showing-its-metals (accessed on 28 
February 2018). 
Öztürk, İlhan, and Ali Acaravcı. (2010). Testing the export-led growth hypothesis: Empirical 
evidence from Turkey. The Journal of Developing Areas 44, no1: 245-254 
151 
 
University of Ghana http://ugspace.ug.edu.gh 
Pak, S. J. and Zdanowicz, J. S. (1994). A Statistical Analysis of the U.S. merchandise Trade Data 
base and its uses in Transfer Pricing Compliance and Enforcement. Tax Management 
Transfer Pricing Report, 3(1):50–57. 
Pak, S. J., Zanakis, S. H., & Zdanowicz, J. S. (2003). Detecting Abnormal Pricing in 
International Trade: The Greece-USA case. Interfaces, 33(2), 54-64. 
Parliament of Ghana (1989). Small-Scale Gold Mining Act (P.N.D.C.L. 218), Parliament of 
Ghana, Accra.  
Patnaik, I., Gupta, A. S., & Shah, A. (2012). Determinants of Trade Misinvoicing. Open 
Economies Review, 23(5), 891-910. 
Pedersen, P.O (2001). “The Freight Transport and Logistical System of Ghana” Working Paper 
Subseries on Globalisation and Economic Restructuring in Africa, no. xii; Centre for 
Development Research (CDR) Working Paper 01.2. Copenhagen. 
Peprah, K. (2019). Cocoa Plant, People and Profit in Ghana. In Theobroma Cacao-Science for 
Sustainability of Global Cocoa Economy. IntechOpen. 
Peralta, S., X. Wauthy, And T. Van Ypersele (2006). Should countries control international 
profit shifting?, Journal of International Economics 68, 24–37 
Peukert, H. (2012). “Mercantilism,” in Handbook of the History of Economic Thought, vol. 11 
of The European Heritage in Economics and the Social Sciences, pp. 93–121, Springer, 
New York, NY, USA 
Piercy, N.F. (1981). “Company Internationalisation: active and reactive exporting, European 
Journal of Marketing, Vol. 15 No. 3, pp. 24-40 
Platform for Collaboration on Tax (2017). ‘A Toolkit for Addressing Difficulties in Accessing 
Comparables Data for Transfer Pricing Analyses’. International Monetary Fund (IMF), 
Organisation for Economic Co-operation and Development (OECD), United Nations 
(UN), World Bank Group (WBG). Retrieved from: https://www.oecd.org/ctp/toolkit-
on- comparability-and-mineral-pricing.pdf. 
Porter, M.E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance, 
Free Press, New York, NY. 
Portes, R., Rey H., and Oh Y. (2001). Information and capital flows: The determinants of 
transactions in financial assets. European Economic Review 45(4), 783–796. 
Quao, E. (2019, June 6) Personal Interview 
Quarmine, W., Haagsma, R., Sakyi-Dawson, O., Asante, F., van Huis, A., and Obeng-Ofori, D. 
(2012). Incentives for Cocoa Bean Production in Ghana: Does Quality Matter? NJAS – 
Wageningen Journal of Life Sciences, 60 – 63 (2012) 7 – 14. 
Qureshi, T.A. and Mahmood, Z. (2016). The Magnitude of Trade Misinvoicing and Resulting 
Revenue Loss in Pakistan. The Lahore Journal of Economics. 21:2, pp. 1-30 
152 
 
University of Ghana http://ugspace.ug.edu.gh 
Rahman, K. (2018). Overview of Corruption and Anti-corruption in Ghana. Transparency 
International. 
Readhead, A. (2016). Transfer Pricing in the Extractive Sector in Ghana. A Case Study. Natural 
Resource Governance Institute. 
Readhead, A. (2016). Preventing Tax Base Erosion in Africa: A Regional Study of Transfer 
Pricing Challenges in the Mining Sector. London: Natural Resources Governance 
Institute. 
Readhead, A. (2018). Monitoring the Value of Mineral Exports: Policy Options for 
Governments, Manitoba, International Institute for Sustainable Development 
Reuter, P. (2011). Draining Development? Controlling Flows of Illicit Funds from Developing 
Countries. World Bank. Retrieved from 
https://elibrary.worldbank.org/doi/pdf/10.1596/978-0-8213-8869-3 
Reuter, P. (2012). Draining Development? Controlling Flows of Illicit Funds from Developing 
Countries. World Bank. 
Reuter, P. (2017). Illicit Financial Flows and Governance: The Importance of Disaggregation. 
Background Paper to the World Development Report 2017. 
Ricardo, D (1817a). Principles of Political Economy and Taxation, G. Bell & sons. 
Ricardo, D. (1817). Principles of political economy in Saffra, P. (Ed.) (1951), The Works and 
Correspondence of David Ricardo, Vol. 1, Cambridge University Press, London 
Roin, J. (2008). ‘Can the Income Tax be Saved? The Promise and Pitfalls of Adopting 
Worldwide Formulary Apportionment’ 61(3) Tax Law Review 169, 174. 
Rose, Andrew and Mark Spiegel (2002). A gravity model of sovereign lending: Trade, default 
and credit. NBER Working Paper Series No. 9285 
Salmona, M., Kaczynski, D., and Smith, T. (2014). Qualitative Research In Finance. Australian 
Journal of Management, 39(1). https://doi.org/10.1177/0312896212469611  
Salomon, M. (2018). 3. Estimating IFFs/TM: Partner Country Method. Illicit Financial Flows via 
Trade Mis-invoicing, 31. 
Salvatore, D. (1998). International Economics, 6th Edition, Prentice-Hall, New Jersey, United 
States of America 
Sarpong, G. (2019, February 4). Ghana lost GHC2bn through collusion of Customs Officers and 
transit traders in 2018. IWatchAfrica. Retrieved from: 
http://iwatchafrica.org/2019/02/04/ghana- lost-gh%C2%A22bn-through-collusion-of-
customs-officers-and-transit-traders-in-2018/ 
Schimanski, C. (2018). Do Multinational Companies Shift Profits out of Developing Countries? 
How Data Availability may hide the evidence (No. 2018/52). WIDER Working Paper. 
153 
 
University of Ghana http://ugspace.ug.edu.gh 
Sen, S. (2010). International trade theory and policy: a review of the literature. Working paper, 
Levy Economics Institute 
Shaxson, N. (2012). Treasure Islands: Tax Havens and the Men who stole the World. Random 
House.  
Shehu, A.Y. (2012). Promoting financial inclusion for effective anti-money laundering and 
counter financing of terrorism (AML/CFT). Crime Law Soc Change 57, 305–323 
https://doi.org/10.1007/s10611-011-9351-0 
Smith, A. (1776). Wealth of Nations, University of Chicago Bookstore 
Suliman, K. M. (2005). The Impact of trade Liberalization on Revenue MobilizationMobilisation 
and stability in Sudan.” African Development Review, 2005, 17: 3, pp. 405-434. 
Tandon, S., & Rao, R. K. (2017). Trade Misinvoicing: What can we Measure? (No. 17/200). 
Tanzi, V. (1987). Quantitative Characteristics of the Tax Systems of Developing countries. In 
The Theory of Taxation for Developing Countries, Newbery D, Stern N (eds). Oxford 
University Press: New York. pp. 205-241 
Tawiah, A. (2018, March 5) Personal interview 
Teera, J. M. and Hudson, J. (2004): “Tax Performance: A Comparative Study. Journal of 
International Development, 16, 785-802.  
Teichmann, F., Maximilian, J., Falker, M.C. (2020). Money laundering through deposit boxes. 
Journal of Money Laundering Control 23: 805–18 
Temiz, D., and Gökmen A. (2010). Analysis of the export and economic growth in Turkey over 
the period of 1950-2009. International Journal of Economic and Administrative Studies 
3, no.5: 123-142. 
The World Bank. Schoot, Allan, P. (2003). Reference guide to Anti-Money Laundering and 
Combating the financing of Terrorism. Washington, DC: The World Bank. 
Thomas Arthur L. (1971). “Transfer Prices of the Multinational Firm: When Will They Be 
Arbitrary?” A Journal of Accounting, Finance and Business Studies 7(1), 40-53. 
https://doi.org/10.1111/j.1467-6281.1971.tb00394.x  
Tosun, M. S. and Abizadeh, S. (2005). Economic Growth and Tax Components: An Analysis of 
Tax Changes in OECD. Applied Economics, 37, pp. 2251-2263 
Tropin, M. J. (2007). ‘ EC Approves Home State Taxation Pilot for SMEs, to Apply Formulary 
Tröster, B., Staritz, C., Grumiller, J. and Maile, F. (2019). Commodity Dependence, Global 
Commodity Chains, Price Volatility and Financialisation: Price-Setting and 
Stabilisation in the Cocoa Sectors in Cote d’Ivoire and Ghana. Working Paper 62. 
Austrian Foundation for Development Research. 
154 
 
University of Ghana http://ugspace.ug.edu.gh 
Turnbull, P.W. (1985). Internationalisation of the firm: a stages process or not?, paper presented 
at a conference on Export Expansion and Market Entry Modes, Dalhousie University, 
Halifax, October 
United Nations (2017). United Nations Practical Manual on Transfer Pricing for Developing 
Countries. 2nd ed. New York: United Nations. Retrieved from: 
https://www.un.org/esa/ffd/wp-content/uploads/2017/04/Manual-TP-2017.pdf 
United Nations Economic Commission for Africa (2015). Illicit Financial Flows: Report of the 
High Level Panel on Illicit Financial Flows from Africa.  
UNODC. (2011). Estimating illicit financial flows resulting from drug trafficking and other 
transnational organized crime. Research report. Vienna 
Vernon, R. (1966). International investment and international trade in the product cycle, 
Quarterly Journal of Economics, May, pp. 190-207. 
Vernon, R. (1971). Sovereignty at Bay, Basic Books, New York, NY 
Vogel, B. and Maillart, J. B. (2020). National and international anti-money laundering law: 
developing the architecture of criminal justice, regulatory law and data protection, The 
Hague: Intersentia. 
Weiner, J. M. (2002) ‘Formulary Apportionment: Possible Solution to the Tax Harmonization 
Dilemma in the European Union?’ Accounting and Tax Periodical 241. 
Welch, L.S. and Luostarinen, R. (1988). Internationalisation: evolution of a concept.  Journal of 
General Management, Vol. 14 No. 2, pp. 34-55. 
Wells, L.T. (1968). “A product life cycle for international trade? Journal of Marketing, Vol. 33, 
July, pp. 1-6. 
Wells, L.T. (1969). Test of a product cycle model of international trade, Quarterly Journal of 
Economics, February, pp. 152-62. 
Whitehouse, D. (2019, October 8). Ghana now Africa’s largest gold producer, but Reforms 
Await. The Africa Report. Retrieved from: 
https://www.theafricareport.com/18245/ghana-now-africas- largest-gold-producer-but-
reforms-await/ 
Williams, T. (2009). An African Success Story: Ghana’s Cocoa Marketing System. IDS Working 
Papers, 2009(318), 1-47. https://doi.org/10.1111/j.2040-0209.2009.00318_2.x 
Wilson, C (1959). The other face of mercantilism, Transactions of the Royal Historical Society, 
vol. 9, pp. 81–101. 
World Bank (2016). The World Bank Group’s Response to Illicit Financial Flows: A 
Stocktaking, Board Report No. 104568 (Washington, DC: The World Bank), Retrieved 
from: http://documents.worldbank.org/curated/en/502341468179035132/pdf/104568-
155 
 
University of Ghana http://ugspace.ug.edu.gh 
BR-SecM2016-0112-IDASecM2016-0071-IFC-SecM2016-00423-MIGA-SecM2016-
0044-Box394878B-PUBLIC-disclosed-4-5-16.pdf  
World Customs Organisation (2018). Illicit Financial Flows via Trade Mis-invoicing. Retrieved 
from: https://www.wcoomd.org/-
/media/wco/public/global/pdf/media/newsroom/reports. 
Yalta, A. Y., & Demir, I. (2010). The Extent of Trade Misinvoicing in Turkey: Did Post-1990 
Policies Matter? Journal of Economic Cooperation and Development, 31(3), 41–66. 
Young, S. (1990). Internationalisation: introduction and overview, International Marketing 
Review, Vol. 7 No. 4, pp. 5-10 
Yu, K., Lu, Z., Stander, J. (2003). Quantile regression: applications and current research areas. 
Journal Of The Royal Statistical Society: Series D (The Statistician),52(3), 331-350.  
https://doi.org/10.1111/1467-9884.00363  
Zdanowicz, J. S. (2009). Trade-based Money Laundering and Terrorist Financing. Review of 
Law & Economics, 5(2), 855-878. 
Zdanowicz, J., Pak, S., & Sullivan, M. (1999). Brazil-United States Trade: Capital Flight through 
Abnormal Pricing. The International Trade Journal, 13(4), 423-443. 
Zhang, W. B. (2008). International Trade Theory: Capital, Knowledge, Economic Structure, 
Money, and Prices Over Time, Springer 
 
 
 
 
 
 
 
 
 
 
 
 
 
156 
 
University of Ghana http://ugspace.ug.edu.gh 
APPENDICES 
APPENDIX 1: RESULTS OF HAUSMAN TESTS 
 
Hausman Test for Overvaluation: Transfer Pricing Risk 
 
Hausman (1978) specification test  
     Coef. 
 Chi-square test value 5.031 
 P-value 0.754 
 
Hausman Test for Overvaluation: Transfer Pricing Rank 
 
Hausman (1978) specification test  
     Coef. 
 Chi-square test value 6.765 
 P-value 0.562 
 
Hausman Test for Undervaluation: Transfer Pricing Risk 
 
Hausman (1978) specification test  
     Coef. 
 Chi-square test value 6.5 
 P-value 0.591 
 
Hausman Test for Undervaluation: Transfer Pricing Rank 
 
Hausman (1978) specification test  
     Coef. 
 Chi-square test value 8.184 
 P-value 0.416 
 
 
 
 
 
 
 
157 
 
University of Ghana http://ugspace.ug.edu.gh 
APPENDIX 2: DESCRIPTIVE STATISTICS 
 
Descriptive Statistics  
 Variable  Obs  Mean  Std. Dev.  Min  Max 
 Overvaluation 356 583283.05 1425684.8 0.005 15655485 
 Undervaluation 463 8322992.9 56931896 0.005 844300000 
 Current Account Balance 756 3580000000 113000000000 -4600000000000 312000000000 
 Corruption 784 58.622 20.02 23 92 
 Inflation 773 3.443 5.148 -1.667 58.933 
 Tax Differential 560 0.019 0.081 -0.08 0.3 
 Interest Rate 496 3.999 7.578 -12.857 41.76 
 GDP Growth 784 2.713 2.742 -10.945 15.816 
 REER 588 95.775 11.834 69.417 129.954 
 Transfer Pricing Risk 596 3.032 0.802 0.787 4.049 
 Transfer Pricing Rank 596 0.638 0.291 0.017 1 
 
 
 
APPENDIX 3: ENDOGENEITY TEST 
---------------------------------------------------------------------------------------------------------------- 
Endogenous-test 1 
Hansen J statistic (overidentification test of all instruments):         0.392 
                                                                       Chi-sq(2) P-val =    0.8218 
-endog- option: 
Endogeneity test of endogenous regressors:                                   1.442 
                                                                       Chi-sq(1) P-val =    0.2298 
Regressors tested:    GDPGrowth 
---------------------------------------------------------------------------------------------------------------- 
Instrumented:         GDPGrowth 
Included instruments: InflationRate CorruptionIndicator taxdiff InterestRate tpr lnCAB REER 
Excluded instruments: L.GDPGrowth L2.GDPGrowth L3.GDPGrowth 
Endogenous 2-test 2 
---------------------------------------------------------------------------------------------------------------- 
158 
 
University of Ghana http://ugspace.ug.edu.gh 
 
Hansen J statistic (overidentification test of all instruments):         4.022 
                                                                       Chi-sq(2) P-val =    0.1339 
-endog- option: 
Endogeneity test of endogenous regressors:                                   0.143 
                                                                       Chi-sq(1) P-val =    0.7057 
Regressors tested:    GDPGrowth 
---------------------------------------------------------------------------------------------------------------- 
Instrumented:         GDPGrowth 
Included instruments: InflationRate CorruptionIndicator taxdiff InterestRate 
TransferPricingRank lnCAB REER 
Excluded instruments: L.GDPGrowth L2.GDPGrowth L3.GDPGrowth 
 
Endogenous 3-test 3 
---------------------------------------------------------------------------------------------------------------- 
 
Hansen J statistic (overidentification test of all instruments):         3.966 
                                                                       Chi-sq(2) P-val =    0.1377 
-endog- option: 
Endogeneity test of endogenous regressors:                                   0.556 
                                                                       Chi-sq(1) P-val =    0.4559 
Regressors tested:    GDPGrowth 
---------------------------------------------------------------------------------------------------------------- 
Instrumented:         GDPGrowth 
Included instruments: InflationRate CorruptionIndicator taxdiff InterestRate 
                      tpr lnCAB REER 
Excluded instruments: L.GDPGrowth L2.GDPGrowth L3.GDPGrowth 
---------------------------------------------------------------------------------------------------------------- 
159 
 
University of Ghana http://ugspace.ug.edu.gh 
Endogenous 4-test 4 
Hansen J statistic (overidentification test of all instruments):         2.647 
                                                                       Chi-sq(2) P-val =    0.2662 
-endog- option: 
Endogeneity test of endogenous regressors:                                   0.564 
                                                                      Chi-sq(1) P-val =    0.4528 
Regressors tested:    GDPGrowth 
---------------------------------------------------------------------------------------------------------------- 
Instrumented:         GDPGrowth 
Included instruments: InflationRate CorruptionIndicator taxdiff InterestRate 
TransferPricingRank lnCAB REER 
Excluded instruments: L.GDPGrowth L2.GDPGrowth L3.GDPGrowth 
 
160