Experts Report on Reparations

Experts Report on Reparations
for
The International Court of Justice
* * *
Case Concerning Armed Activities on
the Territory of the Congo
The Democratic Republic of the Congo v. Uganda
19 December 2020
1
Table of Contents
Introduction 2
Report 1. Loss of Life : Direct Deaths 7
Report 2. Conflict-Related Excess Mortality 19
Report 3. Quantum Recommended Amounts: Human Lives and Property Damage 42
Report 4. Exploitation of Natural Resources 70
2
Introduction
1. This report presents expert opinions regarding estimates of reparations owed to the Democratic
Republic of the Congo by Uganda for the injury caused as a result of the breach by Uganda of its
international obligations between 1998 and 2003, as determined by the International Court of
Justice in its 19 December 2005 Judgment.
Terms of Reference
2. Under the Terms of Reference (TOR) for this report, damages fall into three categories as
described below
:
2.1 I. Loss of human life
(a) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the global
estimate of the lives lost among the civilian population (broken down by manner of death) due
to the armed conflict on the territory of the Democratic Republic of the Congo in the relevant
period?
(b) What was, according to the prevailing practice in the Democratic Republic of the Congo in
terms of loss of human life during the period in question, the scale of compensation due for
the loss of individual human life?
2.2 II. Loss of natural resources
(a) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the
approximate quantity of natural resources, such as gold, diamond, coltan and timber,
unlawfully exploited during the occupation by Ugandan armed forces of the district of Ituri in
the relevant period?
(b) Based on the answer to the question above, what is the valuation of the damage suffered
by the Democratic Republic of the Congo for the unlawful exploitation of natural resources,
such as gold, diamond, coltan and timber, during the occupation by Ugandan armed forces of
the district of Ituri?
(c) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the
approximate quantity of natural resources, such as gold, diamond, coltan and timber,
plundered and exploited by Ugandan armed forces in the Democratic Republic of the Congo,
except for the district of Ituri, and what is the valuation of those resources?
2.3 III. Property damage
(a) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the
approximate number and type of properties damaged or destroyed by Ugandan armed forces
in the relevant period in the district of Ituri and in June 2000 in Kisangani?
3
(b) What is the approximate cost of rebuilding the kind of schools, hospitals and private
dwellings destroyed in the district of Ituri and in Kisangani?
Appointment of independent experts
3. Four independent experts were appointed by Order of the Court for the purposes of
determining reparations:
4. Dr Debarati Guha Sapir (PhD, Epidemiology) is from India where she studied at the University of
Calcutta. She completed her post graduate studies at the Schools of Public Health of Johns
Hopkins University, Baltimore and the Université de Louvain medical faculty. She
became Director of the Centre for Research for Epidemiology of Disasters (CRED) University of
Louvain in 1994 and Professor in 1996. She has specialised on public health and epidemiology in
humanitarian settings following natural disasters and civil conflict, often focusing on issues
related to mortality and disease control. Debarati is widely published in respected scientific
journals including The Lancet, Science Nature scientific reports as well as in the international
press. She founded two unique global data bases on conflicts (CEDAT) and disasters (EMDAT)
which underpins some of the important international reports such those by the IPCC and UN.
She won the Peter Safar Award for Disaster Medicine from the World Association of Disaster and
Emergency Medicine in 2000 and is member of the Royal Academy of Medicine of Belgium.
5. Dr Michael Nest (PhD, Politics) is a consultant focusing on governance and anticorruption issues
in the natural resources sector, including extensive research on ‘coltan’ (tantalite), advising
Central African governments on improving governance over mineral supply chains, and
improving regulation of legal artisanal and small-scale mining. Past clients include OECD, GIZ,
Transparency International, U4 Anti-Corruption Resource Center, Timor-Leste’s Anti-Corruption
Agency, Ghana’s Environmental Protection Agency, and the Independent Commission Against
Corruption in Sydney, Australia. His PhD focused on how mining interests shaped politics in the
DRC during the war from 1998 to 2003.
6. Mr Geoffrey Senogles (FCA, BA Hons) is a Welsh Chartered Accountant who is a partner in
Senogles & Co, Chartered Accountants, Switzerland. He regularly acts as a quantum expert
witness in international arbitrations and has testified on around 50 occasions in court and before
tribunals and overall, has acted in many hundreds of individual cases. Since 1995, he has acted
for claimants, respondents and also as a tribunal-appointed expert. After qualifying in the early-
1990s while in practice in United Kingdom, he moved to Geneva to work on staff at the United
Nations Compensation Commission between 2000 and 2003. Details on the firm are available
at: www.senogles.com
7. Dr Henrik Urdal (PhD, Political Science) is Director and Research Professor at the Peace Research
Institute Oslo (PRIO). He is a former Research Fellow with the International Security Program
at Harvard Kennedy School. Urdal’s work on the impact of demographic and environmental
change on armed conflict, and on the demographic consequences of armed conflict has been
published in leading academic journals. He has been a consultant for organizations like the
World Bank, United Nations, and USAID. Urdal has worked extensively on global trends in armed
conflict as past Director for the PRIO Conflict Trends project.
4
Structure of this document
8. This document contains four reports, each one presenting an expert opinion regarding
reparation estimates on the topics described in the terms of reference. Each report is distinct,
with a separate author and author signature on the final page.
9. However, footnotes, paragraph numbers and page numbers run consecutively throughout this
entire combined document and are not distinct to each report.
9.1 Report 1 (Henrik Urdal) estimates loss of life in terms of conflict-related direct deaths,
including both intentional killings of civilians and civilians who were unintended victims of
violence. Deaths of military personnel are not included as they were not part the TOR.
9.2 Report 2 (Debarati Guha-Sapir) estimates civilian deaths in excess to normal mortality rates
that can be attributed to the conflict. It does not include intentional (direct or indirect)
civilian deaths (these are addressed in Report 1).
9.3 Report 3 (Geoffrey Senogles) estimates the quantum of recommended reparation amounts
for human deaths, as well as damage to, and looting of, property.
9.4 Report 4 (Michael Nest) estimates reparations related to exploitation of natural resources.
Summary of estimated damages
10. Table A is a summary of estimated reparations. Details on the numbers of direct deaths,
intentional (1a) and direct deaths, collateral (1b) can be found in Report 1; with detailed
numbers of excess civilian deaths and associated reparation calculations found in Report 2; the
reparation figures shown for other violence, property damage and looting are in Report 3; and
the reparation figures shown for exploitation of natural resources are in Report 4.
Table A: Summary of estimated reparations, USD
Ituri Outside Ituri Total
1(a). Direct deaths, intentional 130,230,000.0 206,580,000.0 336,810,000.0
1(b). Direct deaths, collateral 21,420,000.0 30,120,000.0 51,540,000.0
2. Excess civilian deaths 5,860,020,000.0 68,521,605,000.0 74,381,625,000.0
3. Other violence, property damage
and looting
24,866,906.0 115,257,956.0 140,124,862.0
4. Exploitation of natural resources 38,986,151.6 16,823,389.6 55,809,541.2
Total 6,075,523,057.6 68,890,386,345.6 74,965,909,403.2
11. The Court’s findings as to reparation totals in respect of lines 1(a), 1(b) and 2 (in Table A above)
will be calculated by the Court by adopting their own findings as to the appropriate numbers of
deaths or other acts of violence and then multiplying each figure by the relevant recommended
individual dollar amount, as found by the Court, and taken from Table B below.
5
12. For the avoidance of any doubt, we highlight that it remains entirely for the Court to make its
own legal findings on these matters and hence the Court will derive its own computations of any
awards of reparations at its own discretion. The figures we present in this report are for
consideration by the Court in the context of its own legal findings.
Table B: Estimated recommended reparations per person, per event or act.
Event or Act Amount, USD
A) Human lives lost
Deaths/injuries resulting from acts of violence deliberately targeted at civilian populations 30,000
Deaths/injuries not resulting from violence targeted at civilian populations but rather, as
collateral victims
15,000
B) Injuries and mutilations
Injury resulting from acts of violence deliberately targeted at civilian populations
Based on Congolese court awards:
- Serious injury 3,500
- Minor injury 150
Based on Congolese ordinary courts:
- Minor injury 100
Injury not resulting from violence targeted at civilian populations but as collateral victims
Based on Congolese court awards:
- Ituri: serious injury 3,500
- Ituri: minor injury 150
Based on Congolese ordinary courts:
- Eastern Congo, Ituri, Kisangani: minor 100
C) Incidences of rape
Based on Congolese court awards:
(* “Simple” is the term used by DRC courts) - “Simple” rape* 5,000
- Aggravated rape 5,000
D) Child soldiers
Based on a figure deemed reasonable by DRC: 10,000
E) Population flight and displacement
Based on figures deemed reasonable by DRC:
- Ituri 300
- Eastern Congo and Kisangani 100
6
7
Report 1
Loss of Life: Conflict Deaths
Dr. Henrik Urdal
(Oslo, 19th December 2020)
8
Summary of loss: quantity and value
13. In accordance with the court order dated 12 October 2020, this report provides ‘...an expert
opinion [...] regarding [...] the loss of human life (in particular, the global estimate of the lives
lost among the civilian population due to armed conflict on the territory of the DRC and the
scale of compensation due’. This section of the report deals exclusively with lives lost as a direct
result of the armed conflict, and covers armed conflict events that took place in the Democratic
Republic of the Congo between 1 August, 1998 and 2 June, 2003.
14. Investigating direct conflict deaths based on the authoritative conflict data collected by the
Uppsala Conflict Data Program (UCDP), the report finds that a total of 28,981 individuals lost
their lives in armed conflict events in the Democratic Republic of the Congo during this time
period. Out of this total number of direct deaths, 14,663 were civilians. Of these civilian victims,
11,227 were killed as a result of deliberately targeted violence and 3,436 were civilian collateral
victims. The estimated total value of civilian lives lost amounts to USD 388,350,000.
Methodological approach and sources
17. Data on armed conflict events form the basis for this report and are collected from the publicly
available source the Uppsala Conflict Data Program, UCDP (https://ucdp.uu.se/). UCDP is hosted
by the University of Uppsala Department of Peace and Conflict Research (DPCR) and is an
independent observatory and collector of armed conflict data globally. It is one of the most
trusted sources of armed conflict data in academia as well as the policy domain, as measured
both by the number of citations to the data in academic works, and references in key policy
reports by international organizations like the UNDP and the World Bank.
18. The UCDP collects data on three types of organized violence, which all apply to armed activity in
the DRC and are in the following collectively understood as armed activity: state-based conflict
(conflict involving the use of armed force between two organised armed groups of which at least
one is the government of a state), non-state conflict (conflict involving the use of armed
force between two organised armed groups, neither of which is the government of a state), and
one-sided violence (the deliberate use of armed force by the government of a state or by a
formally organised group against civilians). For an armed conflict to be registered in the
database it needs to have passed the minimum threshold of at least 25 battle-related deaths in a
calendar year (this could be either military or civilian deaths, killed as a direct result of armed
violence). For definitions of the various types of armed conflict and of battle-related deaths, see
Appendix 2.
19. Specifically, this report utilizes the so-called UCDP Georeferenced Events Dataset (UCDP-GED).
This dataset encompasses all three conflict types described above and contains individual events
within each conflict that are spatially and temporally defined. Note that the three types of
armed violence are mutually exclusive, an event can only be coded in one category. Each event –
defined as an instance of organized violence with at least one fatality – comes with information
on actors, dyad, conflict, geographic location and coordinates, as well as the specific dates on
which the violence took place and fatality estimates (Sundberg & Melander, 2013: 524). For a
more comprehensive definition of armed conflict events, see Appendix 3.
9
20. A representation of these data is provided in Figure 1, in which each circle indicates the number
of conflict deaths by geographical location. Note that Figure 1 displays data points that have, for
presentational purposes, been aggregated to a higher geographical level than the individual
event data points actually represent (these events all have a specific location down to the village
level), and that the figure displays conflict deaths for the entire temporal domain of the dataset
(1989-2019). In contrast, the analysis in this report is based exclusively on data from the period 1
August, 1998 through 2 June, 2003.
Figure 1.1: Armed conflict deaths in the Democratic Republic of the Congo, 1989-2019
Battle deaths estimates
21. The UCDP GED provides three levels of estimates for deaths for each event (for details see the
UCDP-GED codebook, Croicu & Sundberg 2016). These estimates represent an uncertainty
interval ranging from a ‘low estimate’, which provides the most conservative, or cautious,
estimate of deaths that is identified in the sources used; a ‘best estimate’ containing what is
considered to be the most reliable estimate of deaths identified in the sources; and a ‘high
estimate’, representing the highest reliable estimate of deaths identified in the relevant sources.
Even the ‘high estimate’ is considered to be a moderate assessment as the UCDP specifically
avoids including unreasonable claims in the high estimate of fatalities. Generally, UCDP fatality
numbers are conservative. In this report, UCDP ‘best estimates’ are used for all calculations of
lives lost.
Civilian casualties
22. Furthermore, battle deaths can either relate to members of armed groups taking part in combat
between warring parties, or violence against civilians. Civilian deaths can exist in all three
10
categories of violence (Croicu & Sundberg 2016: 27). In the categories state-based and non-state
violence, civilian deaths are considered ‘collateral’ mortality, meaning that victims are
accidentally killed in the fighting between the warring parties. On the contrary, one-sided
violence represents the targeted killing of civilians.
Data collection procedures
23. The UCDP data collection procedures involve the use human coders that manually mine written
sources (Sundberg & Melander, 2013: 525). The process is done in a "two-pass" system, first by
consulting newswire sources for the entire globe, specifically Reuters News, Agence France
Presse (English language version), Associated Press, Xinhua (English language version), and BBC
Monitoring. On the basis of findings from this ‘first-pass’ newswire, local and specialized sources
are consulted in order to code the full range of events. This second-pass level includes NGO
reports, case studies, truth commission reports, historical archives as well as other specialized
sources of information.
24. Data quality in the coding of specific events is further ensured through a procedure in which the
coder first runs through a checklist of consistency and streamlining tests. Following that
procedure, a project manager performs similar tests, as well as controls of the geocoding
through a set routine of visualization. Third, algorithms in PHP and Python are run on the data to
check consistency across unique conflict IDs, coordinates, fatality counts, and other information.
These procedures generally ensure high data reliability.
Data availability and limitations
25. While the UCDP armed conflict data meet the highest academic standards, there are important
limitations. First and foremost, relying on written secondary sources, and primarily newswires,
there is a considerable potential for media and urban biases. As noted in the UCDP codebook
(Croicu & Sundberg 2016: 12) ‘media reporting is not consistent across time or space […].
Changing managerial focuses, different organizational structures (such as field office locations),
as well as different resource distributions and allocations (such as, for example, the restructuring
of BBC Monitoring in the early 2010s) make media reporting quality and quantity vastly different
over various periods and over different areas.’
26. The news organizations also follow different reporting practises, which have changed over time.
Specifically, in some geographical contexts during particular time periods, reporting may have
taken the form of summary reports covering larger areas over a longer time period, which may
hamper the ability of coders to accurately identify specific events. These limitations may be
particularly challenging when studying variation in patterns across many countries over a long
period of time, but are considered less of a problem when studying one country during a limited
time period, such as the DRC over the 1998-2003 period. However, unequal access of reporters
and other independent observers in various parts of the country during various periods of the
armed conflict is likely to have led to the underreporting of armed activity that could have
qualified for inclusion in the UCDP GED event database.
27. Furthermore, there are inherent limitations resulting from the way that the UCDP coding
scheme is constructed. Specifically, the dataset does not include all possible instances of armed
11
activity that may be considered of interest to this case resulting from three key delimiting
factors. First, the dataset only includes events associated with armed conflicts in which at least
25 battle deaths have occurred within a calendar year. This implies that low-intensity violent
conflicts that do not meet this intensity criterion will not be included. However, when a conflict
has passed this threshold in a given year and thus is included in the dataset, all armed activity
events with at least one battle death occurring in preceding years will be included, also in years
in which the conflict in question does not pass the 25 battle deaths threshold.
28. Second, the insistence that actors engaged in armed conflict shall be organized and identifiable
implies that conflicts leading to a significant number of deaths, but where one or more of the
involved armed groups display a low degree of organization for instance by not having a formally
declared name, may not be included in the armed conflict dataset. Third, the dataset only
contains events in which it is possible to establish that there have been fatalities. Events for
which casualty estimates cannot be established are by definition not included.
Key assumptions
29. Biases and strict inclusion criteria discussed above make it more likely that there are deaths
related to relevant armed activities in the DRC in the period 1 August, 1998 through 2 June, 2003
that are not included in the aggregate number of deaths presented in this report, than the
opposite theoretical possibility of an overestimation of casualties. However, I have no data
allowing for an assessment of what this potential underestimation of direct conflict deaths may
amount to. Note that while the UCDP does provide a ‘high estimate’, this estimate is reflecting
uncertainties in the number of casualties in the reports forming the basis for the coding of
specific events. This ‘high estimate’ is hence not taking into account possible sources of
undercounting stemming from the non-registration of potentially relevant events and should for
that reason also be considered a moderate estimate of direct deaths.
30. The factors discussed above together suggest that all UCDP estimates, and specifically the ‘best
estimates’ forming the basis for all estimates in this report, can be assumed to be conservative,
or cautious estimates.
31. Furthermore, I have no information of the demographic characteristics of those killed directly in
armed conflict. We can assume based on knowledge about the armed organizations and on
mortality statistics from other conflict settings that the vast majority of deaths among actors in
armed activities as are men, with the largest age group being around 20-30 years. For the
civilians killed directly in armed activities, I have no general knowledge about their demographic
characteristics.
32. Direct conflict deaths are separated between members of armed groups who lose their life in
the fighting (‘military deaths’) and lives lost among the civilian population not taking part
directly in the armed activities (‘civilian deaths’). We are assuming that civilians killed in ‘statebased
conflicts’ and in ‘non-state conflicts’ are collateral damage, however it is possible that
some civilians killed in such events may also have been directly targeted during the fighting. As
such, ‘targeted civilian deaths’ may be underestimated. All civilians killed in ‘one-sided violence’
are considered ‘targeted civilian deaths’.
12
33. In the coding of deaths in armed conflicts, information may not always allow for a precise coding
of whether deaths are occur among ‘civilians’ or ‘military’. This could result from the source not
being sufficiently detailed, but it could also reflect an ambiguity in the distinction between
‘military’ and ‘civilian’. In particular, young men of ‘military age’ may in some contexts be
considered legitimate military targets even when they are not directly involved in the fighting,
nor wearing uniform or bearing arms. While we are unable to separate empirically between
these two sources, a substantial number of deaths are recorded as ‘unknown’ on the
civil/military distinction. Information on civilian vs military deaths is only coded by the UCDP for
‘best estimates’, and not for ‘low estimates and ‘high estimates’.
34. Numbers of conflict deaths are not broken down on conflict actor in the following analysis.
Conflict deaths: Direct conflict mortality
35. Table 1 includes all deaths recorded for all types of armed conflict (‘state-based’, ‘non-state’ and
‘one-sided’) within the territory of the Democratic Republic of the Congo for the entire period of
1st August 1998 through 2nd June 2003. The best estimate for total direct conflict deaths
recorded by the UCDP for this period is 28,981 (with uncertainty ranging from a ‘low estimate’ of
27,817 deaths to a ‘high estimate’ of 50,836. A total of 603 events are recorded having at least
one person killed in the DRC during this time period.
36. Out of the 28,981 direct deaths, 14,663 deaths occurred among civilians, while 6,494 deaths
occurred among armed actors and are defined as ‘military deaths’. For a total of 7,824 deaths
we do not have information on whether the victims are civilian or military.
Table 1.1: Best estimate of direct conflict deaths by year in the Democratic Republic of the Congo, 1st
August 1998 through 2nd June 2003
Year Civilian deaths Military deaths Unknown Total
1998 3,729 2,117 1,048 6,894
1999 3,462 2,130 1,953 7,545
2000 702 1,218 1,132 3,052
2001 474 66 801 1,341
2002 4,584 925 2,399 7,908
2003 1,712 38 491 2,241
Total 14,663 6,494 7,824 28,981
37. Table 2 displays that of the 14,663 civilian deaths in the whole of the DRC 1st August 1998
through 2nd June 2003, 11,227 were targeted civilian deaths. The number of targeted civilian
deaths for the Ituri province separately during the same period is 4,341. Civilians victims defined
as collateral damage total 3,436 for the DRC as a whole, and to 1,428 for the Ituri province. A
total of 111 events are recorded having at least one person killed in the Ituri province during this
time period.
13
Table 1.2 Civilian deaths categorized as ‘targeted civilians’ and ‘collateral damage’:
Category of civilian deaths Geographic domain
All DRC Ituri province
Targeted civilian deaths 11,227 4,341
Collateral victims 3,436 1,428
38. As shown in Table 3, in Ituri province during the period, a total of 10,398 people were recorded
killed in direct battle, of which 5,769 were civilians, 1,036 were military, and 3,593 were
‘unknown’. The estimate for total number of killed in the Ituri province range from 10,1770 (low
estimate) to 14,752 (high estimate).
39. A total of 23 events across the DRC are recorded as having involved troops of the Government of
Uganda as one of the actors. The majority of these events (20) took place in the province of
North Kivu. In these events, a total of 211 people were reported killed, of which 179 were
military deaths (total for all sides of the conflict, of which 3 were reported to be Ugandan
soldiers). A total of 32 civilian deaths are reported across these 23 events.
Table 1.3 Best estimate of direct conflict deaths in Ituri province and in conflict events involving
troops of the Government of Uganda (in the whole of the DRC), 1st August 1998 through 2nd June
2003
Ituri province Conflicts involving troops of
the Government of Uganda
Civilian deaths 5,769 32
Military deaths 1,036 179
Unknown 3,593 0
Total 10,398 211
Compensation amount for the loss of lives
40. Based on the recommended per capita compensation from the table in paragraph 100, Report 3,
and the estimated number of civilians killed as a result of deliberately targeted violence (11,227)
and the number of civilian collateral victims (3,436), the total estimated value of civilian lives lost
amount to USD 388,350,000 (USD 336,810,000 and USD 51,540,000 respectively).
14
Appendices
Appendix 1.1: Sources cited
Sundberg, Ralph & Erik Melander (2013) ‘Introducing the UCDP Georeferenced Event Dataset’,
Journal of Peace Research 50(4): 523-532.
Croicu, Mihai & Ralph Sundberg (2016) ‘UCDP GED Codebook version 5.0’. Department of Peace and
Conflict Research, Uppsala University. https://ucdp.uu.se/downloads/ged/ucdp-ged-50-
codebook.pdf
15
Appendix 1.2: Definitions and categories of armed conflict
For further UCDP definitions please confer the "Definitions" section of the UCDP web page available
at http://www.pcr.uu.se/research/ucdp/definitions/
State-based armed conflict
A state-based armed conflict is a contested incompatibility (the stated - in writing or verbally -
generally incompatible positions) that concerns government and/or territory where the use of
armed force between two parties, of which at least one is the government of a state, results in at
least 25 battle-related deaths in one calendar year.
Non-state conflict
The use of armed force between two organised armed groups, neither of which is the government of
a state, which results in at least 25 battle-related deaths in a year.
One-sided violence
The deliberate use of armed force by the government of a state or by a formally organised group
against civilians which results in at least 25 deaths in a year.
Note that extrajudicial killings in government facilities are excluded from this definition.
Battle-related deaths
Counted as battle-related deaths is the use of armed force between warring parties in a
conflict dyad, be it state-based or non-state, resulting in deaths.
Comment
Typically, battle-related deaths occur in what can be described as "normal" warfare involving the
armed forces of the warring parties. This includes traditional battlefield fighting, guerrilla activities
(e.g. hit-and-run attacks / ambushes) and all kinds of bombardments of military units, cities and
villages etc. The targets are usually the military itself and its installations, or state institutions and
state representatives, but there is often substantial collateral damage in the form of civilians killed in
crossfire, indiscriminate bombings etc. All deaths - military as well as civilian - incurred in such
situations, are counted as battle-related deaths.
The general rule for counting battle-related deaths is moderation. All battle-related deaths are based
on each coder's analysis of the particular conflict. Each battle-related death has to be verified in one
way or another. All figures are disaggregated as much as possible. All figures that are not
trustworthy are disregarded as much as possible in the coding process. Sometimes there are
situations when there is lack of information on disaggregated battle-related deaths. When this
occurs, the coder may rely on sources that provide already calculated figures either for some
particular incidents, or for total number of deaths in the conflict. The UCDP incorporates such death
figures for particular incidents and for an entire armed conflict if they are coherent with the
definition. If they are not, or if there is no independent verification of the figure, it cannot be
accepted.
16
Appendix 1.3: Definition of Armed Conflict Event
An incident where armed force was used by an organised actor against another organized actor, or
against civilians, resulting in at least 1 direct death at a specific location and a specific date”. These
are the specific elements of the definition:
1) Armed force: use of arms in order to promote the parties’ general position in the conflict,
resulting in deaths.
- arms: any material means e.g. manufactured weapons but also sticks, stones, fire, water etc.
2) Organized actor: a government of an independent state, a formally organized group or an
informally organized group according to UCDP criteria:
a. Government of an independent state: The party controlling the capital of a state.
b. Formally organized group: Any non-governmental group of people having announced a name
for their group and using armed force against a government (state-based), another similarly
formalized group (non-state conflict) or unorganized civilians (one-sided violence). The focus
is on armed conflict involving consciously conducted and planned political campaigns rather
than spontaneous violence.
c. Informally organized groups: Any group without an announced name, but which uses armed
force against another similarly organized group (non-state conflict), where the violent activity
indicates a clear pattern of violent incidents that are connected and in which both groups use
armed force against the other
3) Direct death: a death relating to either combat between warring parties or violence against
civilians. UCDP GED provides three estimates for deaths for each event, thus creating an
uncertainty interval:
- a low estimate, containing the most conservative estimate of deaths that is identified in the
source material;
- a best estimate, containing the most reliable estimate of deaths identified in the source
material;
- a high estimate, containing the highest reliable estimate of deaths identified in the source
material. Note that UCDP attempts to distinguish and not include unreasonable claims in the high
estimate of fatalities, and tends to be highly conservative when counting fatalities. In order for an
event to exist, at least one dead needs to be registered in the high, best or low estimate.
4) Specific location: a name and one pair of latitude and longitude coordinates that relate to the
geographical information specified in the source material.
5) Specific date: a specified time period during which armed interactions cause at least 1 fatality. The
normal temporal unit to which an event can be related is a 24- hour day starting at midnight.
- In some cases it is impossible, based on the source material, to reduce the specific date to a
single day as reporting only refers to wider time spans (multipledays) or information on the exact
day is not clear. For these events, a wider time span is provided through the use of the
date_start, date_end and date_prec variables.
17
Appendix 1.4: Signature of expert
Signature of expert
This report has been prepared in accordance with the terms of reference set out by the International
Court of Justice by Henrik Urdal on 19 December 2020:
Signed:
18
19
Report 2
Conflict Related Excess Deaths
Professor (Em.) Debarati Guha-Sapir
(Brussels, 15th December 2020)
20
Table of Contents
1 Conflict History 21
2 Evidence on mortality presented by DR Congo 21
3 My estimations based on publicly available surveys 24
3.1 Background 24
3.2 Methodological overview 25
3.2.1 Data sources 26
3.2.2 Model specifications 26
3.3 Results 27
3.3.1 Applying Memorial percent shares for comparability 27
3.3.2 Regional considerations 28
3.3.3 Demographic distribution of deaths 30
4 Conclusions 32
List of Tables
Table 1: Reparations sought for deaths and injuries presented in the Memorial 24
Table 2: Population and mortality indicators in Ituri, Eastern DRC, and DRC, 1998-2003, and
comparison of final indicators with the DRC Memorial
28
List of Figures
Figure 2.1: Number of deaths in Ituri, Eastern region, and DRC with confidence interval 28
Figure 2.2: Reported CDR by survey location 29
Figure 2.3: Distribution of crude death rate, by province, from available surveys with mortality
recall period between 1998 and 2003
29
Figure 2.4: Share of expected, indirect and direct excess deaths within total observed deaths,
DRC, Eastern Region, and Ituri, 1998-2003
30
Figure 2.5: Ratio of civilians killed per combatant killed in five historic wars 31
21
Conflict History
41. Armed activities in Ituri and the Eastern provinces played a significant role in notching up
civilian casualties. Since August 1998, major events occurred frequently in that region and
at least six Security Council (SC) or other UN resolutions were filed related to the fighting
until 2003. By November 1999, Security Council resolution S/RES/1279 established an UN
Peace Keeping Mission in the Democratic Republic of Congo (MONUC) in the country.
Since then, UN Security Council and UN Commission for Human Rights deployed three
fact-finding missions (May 2000, May 2001, and May 2003). Widespread violations of
basic human rights are documented from Eastern Congo, especially in Ituri. For example,
a detailed report from Human Rights Watch 1 , which described “estimated 3 000 civilians
are brutally massacred when rival militias clashed in the hospital town of Nyankunde, Ituri
district. The event marked the largest massacre of the Second Congo War. Annex 1
presents a detailed timeline of the conflict.
Evidence on mortality presented by DR Congo
42. There are three key concepts that guide the estimations of mortality in conflicts and which
underlie the Democratic Republic of the Congo memorial calculations. Deaths in conflicts
are often separated into three relevant groupings: direct deaths, indirect deaths, and
excess deaths. Deaths that are due to active and direct armed hostilities often referred to
as “direct deaths”. These include deaths of combatants and civilians who have been
targeted or deliberate killings of villagers and other civilians in efforts that are part of
conflict strategies such as terrorising, driving away or exacting retribution. The second
group are “indirect deaths” which are those due to absence of essential health care such
as emergency obstetrics, vaccination, and access to markets, or inability to sustain basic
livelihood activities. Excess deaths are deaths observed in a specific period that are over
and above those that could be expected in that period. These estimations depend on
baselines and base populations that are used.
43. DR Congo provided a Memorial of the Democratic Republic of the Congo entitled the Case
Concerning Armed Activities on the Territory of the Congo (Democratic Republic of the
Congo v. Uganda), Second Phase, Question of Reparation. The following summarises the
parts of the document that is relevant to this section. The DR Congo Memorial makes
reference to three mortality studies undertaken by the International Rescue Committee
(IRC), second undertaken by B. Coghlan et al. (2006), and by the Association pour le
Development de la Recherche Appliquée en Sciences Sociales (ADRASS)(Grey literature
report - Lambert and Lohle-Tart).
1 https://www.hrw.org/news/2009/08/21/dr-congo-chronology#_Failed_Peace_E…)
22
44. The IRC and Coghlan studies estimate that 3.9 million deaths occurred between 1998 and
2003 in the Democratic Republic of Congo. Most deaths were from preventable and/or
treatable diseases. Coghlan et al (2006)’s Lancet article “Mortality in the Democratic
Republic of Congo: a nationwide survey” used data from 19 500 households to yield a
crude mortality rate of 2.1 deaths per 1000 per month. They calculated their death rate
based on a 16-month recall period where they asked the households to report all deaths
in the household for the recall period, which they then extrapolated back to 1998.
Coghlan’s crude death rate estimate is 40% higher (2.1) than the average crude death rate
for the sub-Saharan Africa region, which is 1.5 deaths per 1000 per month.
45. In contrast to the results presented by Coghlan et al, ADRASS2 estimated 200 000 deaths
due to conflict. This estimate is nearly 20 times lower than Coghlan’s and was generated
by ADRASS demographers Lambert and Lohle-Tart using census data projections from
1984 – 2005 and UN figures.
46. The evidence from DR Congo recognises the challenges in accurately estimating the
numbers of deaths due to the conflict and assigning attributable responsibilities:
“It is extremely difficult, not to say impossible, to determine accurately the number of
victims and the scale of the material damage arising from Uganda’s invasion of a
significant part of DRC’s territory.” (DR Congo memorial section 2.49, Page 48)
“Examination of records drawn up by the Congolese mission of inquiry does not enable an
exact figure to be determined either. The documents do enable precise identification of a
number of victims, but they merely illustrate examples of the injuries suffered and by no
means do they purport to be exhaustive.” (DR Congo memorial 2.61, page 49)
“In order to overcome these difficulties, another approach may be followed involving
consultation of scientific studies in the fields of epidemiology and demography which have
examined the excess mortality caused by conflict. Such studies allow all the deaths caused
by the war in DRC between 1998 and 2003 to be taken into account. These deaths are not
just those which result from hostilities or atrocities. They may also have other causes, such
as lack of medical care due to healthcare systems being plunged into chaos, for example.
The assessments of the number of excess deaths are made using extremely sophisticated
calculation models, based on projection curves and a range of data. At this stage, however,
the focus will be on the results obtained, rather than the details of the methods and
procedures employed by the studies in question.” (DR Congo memorial Section 2.62, Page
49)
47. These are legitimate concerns, widely acknowledged by the academic and research
community for this and other civil conflicts. Given these estimates, the DR Congo
memorial retained a conservative 10% of the approximately 4.0 million excess deaths (400
000 victims). This position was articulated, saying:
2 http://adrass.net/WordPress/wp-content/uploads/2010/12/ Surmortalite_en_RDC_1998_ 2004.pdf
23
“Given the caution which should be observed within judicial proceedings, the DRC
considers it reasonable, in this context, to rely on a minimum estimate of 400,000 victims,
that is, one tenth of the IRC figure which emerges from studies published in the most
renowned scientific journals, including The Lancet.”.” (DR Congo Memorial, section 2.70,
page 51)
48. They attributed 45% of these deaths (180 000) to armed action related to Ugandan
incursion and “deemed a consequence of the invasion of a substantial part of Congolese
territory by Uganda.” Combined with the 2000 Congolese armed forces deaths estimated
from survivor interviews and administrative documents, the total number of lives for
which reparations are sought by DR Congo is 182 000.
“There were also the soldiers and officers of the Congolese armed forces (FAC) who died
in the fighting with the Ugandan army or the rebel movements that it supported. The DRC
discussed the estimated 2 000 deaths in the FAC in Chapter 2 of this Memorial” (DR Congo
Memorial, section 7.14, page 185)
49. While the specific details of the methods and reasoning behind these numbers are not
readily available, it is understood that they were reached through analysis of in-country
data collection (interviews and document review) and international research findings. I
note that Coghlan estimates exclude a population of about 5 million due to insecurity and
these are where death rates could plausibly be higher, and hence lead to an
underestimate of the excess deaths.
50. For Ituri specifically, the Congo Memorial estimated a total of 60 000 excess deaths with
67% of those deaths attributable to deliberate violence toward civilians. The remaining
third of deaths are attributed to deterioration of infrastructure and widespread insecurity.
For example, they observe that of over half the operating health centres were closed
(212/400), no surgeons were available, and humanitarian aid was essentially absent. The
estimate of 60 000 comes from the UN Secretary General’s Second Special Report on
MONUC’s and only reflects incidences where the Special Rapporteur and/or other
delegations were able to access the area or interview survivors and witnesses. The
reparations sought for deaths and injuries in the memorial are presented in Table 1.
24
Table 2.1: Reparations sought for deaths and injuries presented in the Memorial
Province /
category
Notes Monetary
valuation
in USD
Total
deaths
Compensatio
n in USD
(‘000)
Ituri –
additional
Deaths attributable to the breach of
Uganda’s obligations as the
occupying power of Ituri from 1998
to 2003; this does not include deaths
resulting from deliberate attacks
against civilians
18 913 20 000 378 260
Ituri –
direct
violence
Average sum of the monetary value
awarded by Congolese courts to
families of victims of intl war crimes
34 000 40 000 1 360 000
Kisangani
(1999-2000)
Victims of the fighting between
Ugandan and Rwandan forces
18 913 920 17 399
Other
regions
Subtract Ituri deaths – 40 000 deaths
attributable to direct acts of violence
against civilians and 20 000 in other
circumstances – and 920 from
Kisangani)
18 913 119 080 2 252 160
Congolese
armed
forces
Died in fighting with the Ugandan
army or the Ugandan supported rebel
movements
18 913 2 000 37 826
Total deaths for which compensation is sought --- 182 000 4 045 646
My estimations (based on publicly available surveys)
51. To reiterate, from my understanding, the numbers of dead due to armed hostilities in the
Memorial Translation are derived from a combination of in-country data collection – including but
not limited to interviews and document review – and international research findings. I have made
estimates for the death tolls that are comparable to the ones presented by the Congo Memorial
but using a larger collection of publicly available mortality surveys. The main aim is to compare
my death estimations to theirs and set out the limitations of both.
Background
52. DR Congo has experienced armed conflict on many fronts since the beginning of 1996. There are
civil unrests and refugee flows on the frontiers with Central African Republic (CAR), Angola, and
Burundi as well as wars with Uganda and Rwanda. The country also dealt with mass internal
25
displacement where villagers move under duress to other parts of the country in order to survive.
Cumulatively, 3.6 million were displaced in 20033 .
53. Most often these internally displaced have very high rates of mortality compared to refugees who,
having crossed an international border, have easier access to international aid. International aid
is particularly important given that DR Congo was one of the least developed countries in the
world, ranking 152 out of 174 in the Human Development Index in 19984.
54. Apart from direct violence, other acute events influence the death tolls in a country undergoing
conflict. Two such events are epidemics and disasters. DRC is plagued by frequent and deadly
epidemics, and seven major outbreaks were notified to the World Health Organisation during the
conflict period. Three of these were cholera (V. Cholerae El Tor strain) and one Marburg
haemorrhagic fever epidemic. Marburg, a filovirus of the same family as Ebola, has an
exceptionally high case fatality rate. For the 1998 – 2000 epidemic, the case fatality rate was 83%,
translating to 83 fatalities out of every 100 cases. During the Second Congo War, there may have
been other epidemics, few of which may have been notified given that surveillance systems in
conflict-affected regions were rendered non-functional.
Methodological Overview
55. Estimations for excess mortality between 1998 – 2003 were made using data from 38 mortality
surveys, which were in the public domain (see 3.2.1). These were readily available in Conflict
Survey Repository – CEDAT which collects conflict mortality surveys from online sources, controls
them for quality, accuracy and completeness and enters the results into a database (Annex 2:
Methodology for entry into CEDAT Repository). A Bayesian hierarchical model was used as the
most stable statistical technique given the paucity of publicly available data. These calculated
estimates were then compared against a baseline value representing the mortality that could be
expected, had the conflict not occurred.
56. Excess mortality is a useful concept that captures both those who died as a direct consequence of
armed violence but also those who were unable to obtain essential life-saving health services such
as emergency obstetrics, key vaccinations, or access fields for family food sustenance. Direct
combat deaths are analysed by Henrik Urdal (PRIO) in a separate section and is only presented as
a part of our total excess death estimations based on data from UCDP (PRIO) source. The excess
mortality values are presented by Ituri, 5 Eastern Region6 and the country as whole, based on the
following formula:
𝐸𝑥𝑐𝑒𝑠𝑠 𝑑𝑒𝑎𝑡ℎ𝑠 = (𝑜𝑏𝑠𝑒𝑟𝑣𝑒𝑑 𝑜𝑟 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑑𝑒𝑎𝑡ℎ𝑠 𝑖𝑛 𝑐𝑜𝑛𝑓𝑙𝑖𝑐𝑡 𝑝𝑒𝑟𝑖𝑜𝑑) −
(𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑑𝑒𝑎𝑡ℎ𝑠 𝑤𝑖𝑡ℎ𝑜𝑢𝑡 𝑐𝑜𝑛𝑓𝑙𝑖𝑐𝑡).
3 (https://www.brookings.edu/wp-content/uploads/2016/07/The-International-…-
Displacement-in-the-DRC-December-2014.pdf.
4 http://hdr.undp.org/sites/default/files/reports/261/hdr_2000_en.pdf
5 During the war 1998-2003 Orientale existed as Uele, Kibali-Ituri and Haut-Congo. Orientale Province was
reconstituted in 1966 from the amalgamation of the Uele, Kibali-Ituri and Haut-Congo provinces. But was
broken up again into the current day provinces in 2015.
6 Bas-Uele, Haut-Katanga, Haut-Lomami, Haut-Uele, Ituri, Bas-Congo, Lualaba, Maniema, Nord Kivu, Nord-
Ubangi, Sud-Kivu, Tanganyika and Tshopo.
26
Data sources
57. There were four main sources for the surveys. A nation-wide mortality survey in early 2004 by
Coghlan et al. collected numbers of deaths from households over past recall over 16 months
preceding the time of the survey (April – July 2004). They then extrapolated back to August 1998
and estimated excess deaths at about 3.5 million using a baseline average mortality rate of the
Sub-Saharan Africa of 1.2/1000/year. Les Roberts and The International Refugee Commission also
undertook three mortality surveys, one in June 2000, a second one in 2001 and the third one in
2002. Additionally, Van Herp et al (MSF-Belgium) conducted mortality surveys in this period.
58. The following information was extracted from the surveys for the analyses: deaths, crude death
rates, recall period, and sample sizes. I approximated the person-months covered by each survey
as the product of the recall period and sample sizes. The response variable was the number of
deaths (events), and the different levels of exposure (person-months) used in the various surveys
was included as the offset variable. All rates were converted to 1000/month. All surveys specified
the region in which they were undertaken and were separated into Eastern and Western regions
to maintain compatibility with the study by Coghlan – the main source for the evidence provided
by DR Congo Memorial. Annex 3 lists the surveys and main characteristics.
Model specifications
59. Faced with paucity of mortality data and no systematic civil registration system for deaths during
those years, a Bayesian approach was used to calculate how many deaths may occurred in the
conflict period. The model separately estimates pooled crude death rates (CDR) from 38 surveys
- 22 surveys in the Eastern Region, 16 in the Western region, and 38 nation-wide. The model
accounts for all parameter uncertainties and additionally, uses information from other surveys to
improve the estimate. 7,8,9 Accordingly, the Posterior CDR (in table 2) is the combined death rate
from other surveys (known as prior and in our case follows a normal distribution) and the
likelihood (number of deaths which follows a Poisson distribution) to obtain one pooled CDR
value.
60. The following formula was used to obtain the excess mortality:
𝐸𝑥𝑐𝑒𝑠𝑠 𝑑𝑒𝑎𝑡ℎ𝑠 =
( 𝑃𝐶𝐷𝑅−𝐵𝐶𝐷𝑅 )
1 000
× 𝑟𝑒𝑐𝑎𝑙𝑙 𝑝𝑒𝑟𝑖𝑜𝑑 × 𝑚𝑒𝑑𝑖𝑎𝑛 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 (1)
Which is essentially described as (deaths estimated from the different surveys during the conflict
period minus the deaths expected if the pre-conflict baseline applied in this period) multiplied by
the (recall period of the survey) multiplied by the (median population of the region).Details are
available in Annex 4: Notes on Bayesian Methodology.
Baselines and base populations
61. Excess mortality calculations are delicate exercises, and two factors are key determinants of the
final result – baselines and base populations. The baseline provides the number of deaths we can
expect in the absence of civil war and reflects deaths due to rampant illnesses, hunger and other
7 Coghlan B, Brennan RJ, Ngoy P, Dofara D, Otto B, Clements M, et al. Mortality in the Democratic Republic of
Congo: a nationwide survey. Lancet. 2006;367(9504):44-51.
8 Lesaffre E LA. Bayesian biostatistics. London: John Wiley; 2012.
9 Sutton AJ, Abrams KR. Bayesian methods in meta-analysis and evidence synthesis. Stat Methods Med Res.
2001;10(4):277-303
27
causes that are simply the result of structural poverty and not due to the conflict. The higher the
baseline value, the lower the excess deaths will be and vice versa. The baseline needs to be
justified as the appropriate choice. The CDR in DR Congo reported by UNICEF in 1997 was 14/1000
per annum translates to 1.2/1000/month was chosen as the baseline for this analysis.10 Although
many surveys use the average Sub-Saharan death rate (1.5/1000/month) as a baseline rate, in this
case, it is best to take a baseline that is closest to the population under study. The UNICEF estimate
is appropriate as it reflects mortality rates from the country itself (rather than a continental
average) and from a period that just precedes the hostilities.
62. Base population is the other key determinant of excess mortality. The bigger the base population
is, the higher the absolute number of excess deaths will be. We obtained the median population
for DR Congo using yearly population projections of the World Bank based on the 1984
Census. It was assumed that the yearly population growth is the same across the provinces
and estimated the monthly population from 1987 to 2003. The base population was
calculated as the monthly median population.
Results
63. In DR Congo as a whole, we would expect about 3.5 million deaths over 1998 – 2003 based on the
UNICEF Crude Death Rate (CDR) of 1.2 for a median population of nearly 50 million, or about 700
000 deaths/year had there not been conflicts within the country. Instead, our point estimate of
total deaths in DR Congo in this 5-year conflict period is 8.5 million or 1.4 million per year and
nearly 5 million deaths in excess to what we may have expected. As a modelled estimate, I can
say with 95% confidence that a minimum of 3.2 million excess deaths may have resulted in this
period due to armed conflict (Table 2 and Figure 1).
64. Deaths in Eastern Congo, including Ituri, accounted for most of the national excess mortality -
about 3.7 million, or over 70% of the total. As no surveys specific to Ituri were available, the death
rates of the Eastern region were applied to a smaller Ituri base population to obtain excess deaths
at just over 400 000.
3.3.1 Applying Memorial percent shares for comparability
65. To make our estimates roughly comparable to those presented by Congo, we can apply some of
the percentages that they have used in the Memorial. The Congo memorial used 10% of the 4
million Coghlan excess death estimate to arrive at 400 000. They then attributed 45% of 400 000
or 180 000 deaths as the responsibility of Uganda and Uganda-supported forces. To this 2000
deaths among the Congolese forces were added for a total of 182 000 deaths attributable to
Uganda and eligible for compensation. Our total excess deaths for DR Congo and Ituri include
deaths of combatants, as the soldiers would not have died, had there been no conflict and
therefore are part of the excess death estimation.
66. Applying these proportions to my modelled estimate of total excess deaths in DR Congo (4 987
756 deaths over 5 years) (Table 2), comparable numbers of deaths for attribution to Uganda are
224 500 excess deaths for the entire country attributable to Uganda compared to 182 000
estimated by DR Congo (green highlight. Similarly, for Ituri, we arrive at 74 000 excess deaths
compared to Congo Memorial estimate of 60 000 deaths attributable to Ugandan and allied forces
(blue highlight).
10 https://www.unicef.org/sowc99/ sowc99e.pdf
28
Table 2.2: Population and mortality indicators in Ituri, Eastern DRC, and DRC, 1998-2003, and
comparison of final indicators with the DRC Memorial
Indicator Ituri Eastern DRC DRC Mem
Median population 2 833 460 26 232 742 49 734 520
Posterior CDR (no. deaths /1000/month) 3.640399 3.640399 2.9290967
Expected deaths* 197 209 1 825 799 3 461 523
Observed deaths 598 266 5 538 864 8 449 279
Total excess deaths 401 057 3 713 065 4 987 756
[CLs](mil) [0.23, 0.62] [2.1,5.7] [3.2,7.1]
Direct excess deaths 10 389 22 935 28 981
Indirect civilian excess deaths 390 668 3 690 130 4 958 775
Using DRC coefficients from the Memorial
45 % of 10% = 4.5% of total excess deaths
attributed to Ugandan Armed action
18048 167088 224449 182 000a
33% (60000/182000) of DRC 73994 Not applicable Not applicable 60000a
*Based on pre-conflict baseline, 1.2/1000/month. CDR: Crude Death Rate; CLs: Confidence Levels, DRC:
Democratic Republic of the Congo. a ref: section 2.71 and 7.14
Figure 2.1: Number of deaths in Ituri, Eastern region, and DRC, with confidence interval.
3.3.2 Regional considerations
67. The highest death rates (over 4 deaths/month/1000) are concentrated in the Eastern Region,
where civil war was raging in this period (Figure 4). Two of the highest were in Kalemia, a port
town on Lake Tanganyika and distribution centre for high value minerals such as zinc, cobalt, and
tin with CDR of 10/1000/month, and Moba also on Lake Tanganyika and part of the Triangle of
Death with 11/1000/month. In Nov. 2000, the Congolese Rally for Democracy-Goma (RCD-Goma)
and the Rwandan Patriotic Army (RPA) forces fought in Moba, a small village of 26 000 inhabitants
where families lived by fishing and farming, and exploited gold mines in that zone. According to
29
our statistics, there were about 300 dead every month during the conflict in Moba and about 1600
per month in Kalemia. Although most of the provinces with high excess mortality rates were in
the Eastern region, Equateur Province in the Western region proves to be an exception. A survey
conducted in Basankusu reported a CMR of over 8/1000/month, which is particularly high in
comparison to other Western provinces. Basankusu was situated on the frontlines between the
Western and Eastern parts of the country where conflict spilled over, devastating the town in
August 2000. The high death rate here may be driven by the extraordinarily high child death rate
of 6.6 per 10 000 per day.
Figure 2.2: Reported CDR by survey location
Annex 2 provides an overview of each survey included in our analyses and specifies the setting of
each survey.
Figure 2.3: Distribution of crude death rate*, by province**, from available surveys with mortality
recall period between 1998 and 2003
* not all current provinces provided public information on death rates.
** provinces identified according to the current administrative divisions
30
3.3.3 Demographic distribution of deaths
68. Finally, in recent years, the conflict death burden is increasingly carried by the civilian population
and less so by the combatants. Our results agree with the DR Congo Memorial with respect to this
aspect. Historically, armed combatants accounted for the majority of the casualties in wars
compared to civilian victims. Since WW2, this ratio shifted to civilian death rates increasing and as
civil conflicts increased, eventually civilian death rates were nearly twice that of soldiers (Figure
4).
Figure 2.4: Ratio of civilians killed per combatant killed in five historic wars
69. Here in the DR Congo, the share of direct battle related deaths is vanishingly small. They are deaths
of armed combatants, consequences of violence such as by massacres, shootings, wholesale
execution of civilians, or large scale burning of settlements and is described in greater detail in the
section on combat deaths by Henrik Urdal. By far the majority perish from other less direct causes
such as conflict induced hunger or disease.
70. Most of the civilian direct deaths are likely to be women and children simply because they
represent by far, the majority of the population (current demographic trends imply that about 70
– 80% of the sub-Saharan Africa population are women and children) and hence mostly). Hence,
they are the most exposed. Second, women with small children are unable, in most circumstances,
to flee rapidly or far enough away to survive killing attacks. The lion’s share of the excess deaths
is due to indirect causes including collapse of the health facilities, uncontrolled disease especially
among children, lack of emergency and obstetric care, breakdown of food supply chains and lack
of access to fields11,12,13. The war in Congo and in Ituri is unlikely to present patterns that are
different from those described above. By far the largest burden of the nearly 5 million excess
11 https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-020-01708…
12 Armed conflict and child mortality in Africa: a geospatial analysis, Zachary Wagner , Sam Heft-Neal , Zulfiqar
A Bhutta , Robert E Black , Marshall Burke, Eran Bendavid
13 War or health: a four-decade armed conflict in Iraq Riyadh K Lafta 1, Maha A Al-Nuaimi https://pubmed.
ncbi.nlm.nih.gov/31597450/
31
deaths 3.7 million excess deaths is centred in the Eastern region and 400 000 in Ituri will have
been carried by these non-combatant sub-groups (Figure 5).
Figure 5: Share of expected, indirect and direct excess deaths within total observed deaths, DRC,
Eastern Region, and Ituri, 1998-2003
Conclusions
71. The mortality estimations were based on 38 surveys undertaken in the period between 1998 –
2003. I estimate the death tolls in DR Congo between 1998 – 2003 that can be attributed to the
conflict to be 401 057 in Ituri, 3 713 065 in Eastern region and 4 987 756 for the country as a
whole. These estimates are slightly higher than those presented by the Congo Memorial but not
significantly different. The difference may be due to the 5 million inhabitants in high insecurity
zones who were excluded from the sample of the Coghlan et al study. In conclusion I estimate the
excess indirect civilian due to the conflict to be 4 958 775. This figure does not include direct
military or targeted civilian deaths. The unit costs per life for indirect civilian life lost is USD 15 000
as per estimation of Geoffrey Senogles.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Ituri Eastern DRC
Share of expected and excess deaths in total deaths observed,
DRC, Eastern Region, and Ituri, 1998-2003
Direct.excess.deaths Expected.deaths Indirect.excess.deaths
32
Appendix 2.1: Timeline of relevant events in Eastern Congo 1998-2003
DATE INCIDENT
August 1998 Ugandan armies invade Congo, backing the Congolese rebel group formed
to oust Kabila
1999 100 miners killed in collapse attributed to UPDF pressures to expedite gold
extraction
29 - 30 May 1999 Burned down villages and burned alive elderly and women in the village of
Loda
20 June 1999 25+ people killed in attack on Dhendro by Hema militias and UPDF soldiers
June - Dec 1999 Lendu civilians and Hema killed in attacks
June - Dec 1999 Tens of Hema-Gegere’s killed in Libi and Fataki
6 Aug 1999 Deployment of 90 military observers to DRC S/RES/1258
14 Sept 1999 Several hundred civilians in Bahema-Nord community killed; victims
buried in mass graves
14 Sept 1999 Several tens of civilians, including 15 minors and a number of women
killed in Fataki
July 1999 100+ Hema civilians killed in Musekere using machetes and edged
weapons
August 1999 Ceasefire signed in Lusaka, Zambia but ignored by rebels
30 Nov 1999 Established UN Mission in DRC (MONUC) S/RES/1279
1 Dec 1999 200 + civilians killed in hostilities between UPDF and Hema militias over
control of Bambou and its mines; bodies were mutilated, and the town
looted; bodies thrown into River Chari
Jan 2000 Several hundred Hema killed with edged weapons
24 Feb 2000 Authorised the expansion of MONUS to consist of 5537 military personnel
S/RES/1291
26 Apr 2000 10 deaths, Lendu civilians, killed by Hema militias and UPDF troops
June 2000 Uganda and Rwanda battle in Kisangani, killing roughly 700 individuals; UN
concludes the 2 armies committed war crimes and calls on Uganda and
Rwanda to pay reparations
June 2000 IRC’s 1st mortality survey reports 1.7 million+ have died in the east since
1998 due to the war
11 May 2000 Security Council mission to the DRC on 4-8 May 2000 S/2000/416
January 2001 President Laurent Kabila killed in Kinshasa; son Joseph assumes power
9 - 18 Jan 2001 Hema killed ~60 people, including Lendu and Ngiti civilians in Kotoni and
Irumu
19 Jan 2001 Between 200-250 Lendu, Ngiti, Nande, Bira ethnic groups killed in Bunia
3 Feb 2001 105 killed by Hema militias and UPDF troops
May 2001 Updated survey from IRC finds death toll increased to 2.5 million
29 May 2001 Security Council mission to the Great Lakes region, S/2001/521
Jan 2002 35 Lendu civilians killed in Kobu; mass displacement of villagers into forest
33
Feb 2002 Inter-Congolese rebel groups sign an agreement, but Rwandan-backed
rebels are excluded
26 Jan 2002 100 Lendu killed in a forest after being chased from the village of Datule
9 Feb-24 Apr 2002 Local NGO reported 2 867 civilian deaths; mass killings occurred on 10 Feb
2002 (173 killed), 15 Feb 2002 (120 killed), 21 Feb 2002 (220 killed), 14
Mar 2002, 29 Mar 2002 (109 killed)
11 May 2002 46 reported deaths in village of Walu from eyewitness accounts
13 May 2002 Security Council mission to the Great Lakes region, 2002 S/2002/537
June 2002 At least 27 Lendu people killed in village of Buba in Walendu Pitsi by UPDF
12 June 2002 Lendu civilians killed; 100 Hema civilians killed in revenge
16 Jul 2002 Report of the joint fact-finding mission on the situation in Kisangani
presented to the Council by the High Commissioner for Human Rights
S/2002/764
9 Aug 2002 Tens of civilians killed in Komanda, mostly Hema civilians
Aug 2002 110+ people killed in and around Bunia
14 - 19 Aug 2002 50 civilians killed in attack on Komanda
28 Aug 2002 Lack of Ugandan intervention in targeted attack on town of Mabanga
resulted in 16+ deaths
31 Aug 2002 At least 14 civilians killed in Songolo
September 2002 Estimated 3000 civilians brutally massacred when militias clash in
Nyankunde in Ituri district
11 Oct 2002 320 bodies buried with 69 of them identified
20 Oct 2002 At least 10 Lendu killed
12 - 29 Oct 2002 At least 173 Nande and Pygmy civilians killed around Mambasa
24 Oct 2002 Dozens killed in Walendu Bindi community
5 Nov 2002 14 civilians killed in a night-time attack, including women; tied up and
killed with machetes
20 Nov 2002 Roughly 200 civilians killed in Mongbwalu after mortar attacks
4 Dec 2002 Peace agreements with DRC’s neighbours & MONUC S/RES/1445
24 Jan 2003 Noted with concern the links between conflict and natural resources in
DRC and renewed the mandate of the Panel of Experts on illegal
exploitation of natural resources S/RES/1457
24 Feb 2003 Roughly 260 persons killed with 173 under the age of 18, an additional 70
people missing
4 Mar 2003 Roughly 168 persons killed in attack on UPC military position in Mandro
6 Mar 2003 Civilians killed in Bunia
10 Mar 2003 At least 100 people killed, many of whom were women and children
20 Mar 2003 Condemned massacre and other systematic violations of International
Humanitarian Law and human rights perpetrated in DRC: in particular,
sexual violence against women and girls in warfare S/RES/1468
April 2003 IRC reports that the death toll of Congolese civilians since 1998 has risen
to 3.3 million
34
3 April 2003 408 civilian deaths reported, 80 serious injuries
27 May 2003 Second special report for the Secretary-General on MONUC (S/2003/566)
30 May 2003 Authorised Interim Emergency Multinational Force deployment (IEMF)
S/RES/1484
June 2003 European troops are deployed to Ituri to protect civilians from fighting
between rival ethnic militias following Uganda’s withdrawal, support to
UN peacekeepers in Bunia
References: HRW – DR Congo Chronology (2009); Security Council Resolutions - UN; Secretary-
General’s Reports - UN; Human Rights Council Documents - UN; Selected Other Documents – UN;
Case Concerning Armed Activities on the Territory of the Congo - ICJ
35
Appendix 2.2: Methodology for entry into CEDAT Repository.
72. Evidence has become central for humanitarian decision making, as it is now commonly
agreed that aid must be provided solely in proportion to the needs and on the basis of needs
assessments. Still, reliable epidemiological data from conflict-affected communities are
difficult to acquire in time for effective decisions, as existing health information systems
progressively lose functionality with the onset of conflicts.
73. In the last decade, health and nutrition humanitarian agencies have made substantial
progress in collecting quality data using small-scale surveys. In 2002, a group of academics,
non-governmental organizations, and UN agencies launched the Standardized Monitoring
and Assessment of Relief and Transitions (SMART) methodology. Since then, field agencies
have conducted thousands of surveys. Although the contribution of each survey by itself is
limited by its small sample and the impossibility to extrapolate results to national level, their
aggregation can provide a more stable view of both trends and distributions in a larger
region. These surveys are compiled by, the Centre for Research on the Epidemiology of
Disasters (CRED) – Université Catholique de Louvain of from online searches either from
journals or posted by the field agencies. The process of entering a survey that is eligible is
presented in Figure 1 below.
74. The surveys are controlled for quality using a standard checklist system (fig.2)* which is
applied to each survey.
Source: The Complex Emergency Database: A Global Repository of Small-Scale Surveys on
Nutrition, Health and Mortality, Chiara Altare and Debarati Guha-Sapir, Plos One, October
21, 2014 https://doi.org/10.1371/journal.pone.0109022
*Quality control check list Figure 2 is at the end
36
37
Appendix 2.3. List of surveys and their characteristics
Survey no. CDR Year Location Province N
361 11.5 2000 East Tanganyika 1212
11 10.8 2001 East Tanganyika 2204
73 8.2 2001 West Equateur 11532
21 7.5 2001 East Maniema 1958
351 6.4 2000 East North Kivu 1330
202 6.3 2002 East North Kivu 1066
172 6.2 2002 East Orientale 1902
31 4.9 2001 East South Kivu 1802
162 4.8 2002 East North Kivu 1119
192 4.6 2002 East Bas-Uele 1345
41 4.4 2001 East South Kivu 1778
132 4.2 2002 East Tanganyika 1372
252 3.9 2002 West Kasai-Central 1161
103 3.3 2001 West Katanga 5077
61 3.0 2001 West Sankuru 1288
182 3.0 2002 East Maniema 1712
232 3.0 2002 West Kwango 1064
51 2.8 2001 East Tshopo 2317
262 2.8 2001 West Kasai-Orientale 1199
331 2.7 2000 East South Kivu 1273
341 2.7 2000 East South Kivu 1219
321 2.6 2000 East Orientale 2305
374 2.4 2004 East Eastern region 82646
83 2.4 2001 West Equateur 8331
272 2.3 2002 West Katanga 1381
212 2.2 2002 East Haut-Uele 1309
122 1.9 2002 East South Kivu 1323
222 1.8 2002 West Kinshasa 1523
384 1.8 2004 West Western region 36732
93 1.8 2001 West Bas Congo 4491
282 1.7 2002 West Katanga 1019
242 1.4 2002 West Kongo 1232
302 1.4 2002 West Kwango 1278
292 1.2 2002 West Kasia-Orientale 1653
113 1.2 2002 West Bandundu 6172
152 0.9 2002 East North Kivu 895
312 0.6 2001 West Equateur 1407
142 0.4 2002 East North Kivu 1373
Sources: 1 Roberts L. Mortality in eastern DRC: results from five mortality surveys. New York:
International Rescue Committee, 2000. Roberts L, Belyadoumi F, Cobey L, et al. Mortality in the
eastern Democratic Republic of Congo: results from 11 mortality surveys. New York: International
Rescue Committee, 2001
38
2 Roberts L, Zantop M. Elevated mortality associated with armed conflict–Democratic Republic of
Congo, 2002. CDC Morbidity and Mortality Weekly Report 2003; 52: 469-71
3 Van Herp M, Parqué V, Rackley E, Ford N. Mortality, violence and lack of access to healthcare in the
Democratic Republic of Congo. Disasters. 2003;27(2):141-53.
4 Coghlan B, Brennan RJ, Ngoy P, et al. Mortality in the Democratic Republic of Congo: a nationwide
survey. Lancet 2006; 367:
39
Appendix 2.4: Bayesian methodology
75. I constructed a hierarchical Bayesian model to estimate separately a pooled CDR for the Eastern
(22 surveys) Western region (16 surveys) and nation-wide (38 surveys). This Bayesian approach
accounts for all parameter uncertainties and in addition, borrows strength from other surveys to
improve the estimate14,15,16. So the Posterior CDR (in table 2) is the pooled value of information
from other surveys (we assume that our prior information follows a normal distribution and the
likelihood which is number of deaths follows a Poisson distribution) to obtain one pooled CDR
value.
76. To obtain the excess mortality we used the following formula:
𝐸𝑥𝑐𝑒𝑠𝑠 𝑚𝑜𝑟𝑡𝑎𝑙𝑖𝑡𝑦 =
( 𝑃𝐶𝐷𝑅−𝐵𝐶𝐷𝑅 )
1 000
× 𝑟𝑒𝑐𝑎𝑙𝑙 𝑝𝑒𝑟𝑖𝑜𝑑 × 𝑚𝑒𝑑𝑖𝑎𝑛 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 (1)
77. Where 𝑃𝐶𝐷𝑅 the posterior crude death rate per 1000 per month and 𝐵𝐶𝐷𝑅 is the Baseline crude
death rate per 1000 per month. The number of deaths, 𝑦𝑖 ( 𝑖 = 1,2 … , 𝑛 ), was modelled using a
Poisson distribution (see equations below), where is the number of surveys. In this analysis 𝑛
takes the value 22 (surveys conducted in provinces in Eastern DRC) or 16 (surveys conducted in
provinces in Western DRC). In statistical notation
𝑦𝑖 = 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (𝑙𝑎𝑚𝑏𝑑𝑎𝑖 ) (2)
log(𝑙𝑎𝑚𝑏𝑑𝑎𝑖 ) = 𝑡ℎ𝑒𝑡𝑎𝑖 + log (𝑝𝑑𝑖 ) (3)
78. Where 𝑦𝑖 is the observed number of deaths in survey 𝑖 and 𝑝𝑑𝑖 is the person-months in survey 𝑖.
The log-transformed observed number of deaths is assumed to follow a normal distribution
(equation 2). The parameter of interest, 𝑡ℎ𝑒𝑡𝑎𝑖 ~ 𝑁𝑜𝑟𝑚𝑎𝑙( 𝑎𝑙𝑝ℎ𝑎, 𝑏𝑒𝑡𝑎 ) where 𝑎𝑙𝑝ℎ𝑎 is the
common posterior crude death rate and 𝑏𝑒𝑡𝑎 is the between-survey variance. Furthermore, I
assign a non-informative normal prior to 𝑎𝑙𝑝ℎ𝑎 ~ 𝑁𝑜𝑟𝑚𝑎𝑙(0,0.000001) and non-informative
gamma prior 𝑏𝑒𝑡𝑎 ~ 𝑔𝑎𝑚𝑚𝑎(0.001,0.001).
79. I used Gibbs sampler to simulate draws from the posterior distribution of the Poisson model, I ran
1000 000 iterations with a burn-in length of 500 000 using three Markov Chain Monte Carlo
(MCMC) with different initial starting values. I checked the convergence of my estimate by visual
inspection of the trace plots.
14 Coghlan B, Brennan RJ, Ngoy P, Dofara D, Otto B, Clements M, et al. Mortality in the Democratic Republic of
Congo: a nationwide survey. Lancet. 2006;367(9504):44-51.
15 Lesaffre E LA. Bayesian biostatistics. London: John Wiley; 2012.
16 Sutton AJ, Abrams KR. Bayesian methods in meta-analysis and evidence synthesis. Stat Methods Med Res.
2001;10(4):277-303
40
Appendix 2.5: Signature of Expert
Signature of expert
This report has been prepared in accordance with the terms of reference set out by the International
Court of Justice by Debarati Guha Sapir on 19 December 2020:
Signed:
41
42
Report 3
Quantum Recommended Amounts:
Human Lives and Property Damage
Geoffrey Senogles
(Nyon, 19th December 2020)
43
SECTION: Injury to persons
80. In their Memorial17, the DRC asserts five claim categories for losses in respect of non-pecuniary
losses caused by injury to persons.
81. These five claim categories and their amounts are summarised in the table below:
DRC Memorial
paras.
Claim amount
USD
A) Human lives lost 7.11-7.15 4,045,646,000
B) Injuries and mutilations 7.16-7.21 54,464,000
C) Incidences of rape 7.22-7.25 33,458,000
D) Child soldiers 7.26-7.28 30,000,000
E) Population flight and displacement 7.29-7.32 186,853,800
4,350,421,80018
82. In this section, for the assistance of the Court I cover each claim category firstly by providing a
synopsis of the basis and evidence provided for the amounts claimed, before secondly going on
to provide my opinion on recommended amounts of compensation. I regard it as appropriate to
set out and comment on, at least briefly, the ways in which the DRC claim amounts have been
derived.
A) Human lives lost – CLAIM: USD 4,045,646,000
83. The majority of the DRC claim value in respect of the overall injury to persons, is made up of the
category human lives lost. This category represent 92% of the Injury to Persons total19.
84. The DRC makes claims for lives lost or injuries suffered in two categories, based on two
differentiated causes of death or injury. These are as follows:
a. Deaths or injuries resulting from acts of violence deliberately targeted at civilian populations20;
b. Deaths/injuries not resulting from violence targeted at civilian populations but, rather, as
collateral victims21.
85. The more aggravated circumstances of an alleged deliberate targetting of civilians, leads the DRC
to attach a higher monetary amount on the individual claim for each death/injury so caused,
when compared to the analagous individual amount in the (somewhat) less grevious
circumstances where a civilian was killed or injured even though not specifically targetted in the
17 DRC Memorial, paras. 7.05-7.32
18 I note that the total claim figure stated in the DRC Memorial appears to arithmetically inaccurate
(overstated). Total claim for Injury to Persons is stated at USD 4,409,108,839 [DRC Memorial, para 7.64].
19 Calculated as: USD 4,045,646,000 / USD 4,350,421,800 = 92%.
20 DRC Memorial, para. 7.08.
21 DRC Memorial, para. 7.09.
44
attack.
86. The DRC’s claim computations are based on the following individual amounts:
86.1 Deaths/injuries resulting from acts of violence deliberately targeted at civilian
populations:
- a flat rate based on Congolese court awards.
- an individual claim amount of USD 34,00022.
86.2 Deaths/injuries not resulting from violence targeted at civilian populations but
rather, as collateral victims:
- a flat rate based on future earning potential
- an individual claim amount of USD 18,91323.
87. Supporting evidence provided by the DRC for the USD 34,000 claimed amount is stated to be in
the two documents (available in the annexes in their original French only) that record written
judgements of two Congolese military courts. Thus, those documents would be expected to
evidence the stated range of court awarded compensation amounts; stated as ranging between
USD 5,000 and USD 100,00024. One of these court documents was published in Kinshasa but
refers to actions in Ituri25 and the other court is located in Bunia26.
88. My review of the two documents reveals that neither of the extracts provided is complete and
neither contains the amounts of compensation awarded by the two courts. Hence, it is not clear
to me how these documents evidence, as they are asserted to do, the amount of USD 34,000 per
individual and it therefore follows that, in my opinion, the individual flat rate amount claimed
has not been supported by clear documentary evidence. Thus, I have no evidentiary basis in the
record on which to measure the extent to which this figure is robust, reliable and reasonable.
89. As to the flat rate individual amount of USD 18,913 (above) claimed in respect of deaths/injuries
of relevant collateral victims27, this figure is derived by the DRC using an averaging methodology
assuming that each such victim has incurred a loss of future income earnings equivalent to the
average Congolese person, of average life expectancy, of average earning capacity and
opportunity, of average age (27 years) within the cohort of victims according to the victim
identification forms completed and filed.
90. In principle, this methodology is not unreasonable but its application in this instance contains
several mattters of detail that are open to question.
22 DRC Memorial, para. 7.12.
23 DRC Memorial, para. 7.09.
24 DRC Memorial, paras. 7.11-7.12.
25 DRC Responses to questions of the Court, Annexe 10.1.
26 DRC Responses to questions of the Court, Annexe 10.2.
27 Collateral victims who were located in Ituri (DRC Memorial, paras 7.30 and 3.41), in Kisangani (paras. 7.24
and 4.47) and elsewhere in Eastern Congo (paras. 7.24 and 2.83).
45
91. For instance, the following comments can be made:
91.1 The victim identification forms made available in the DRC evidence, do not facilitate a
comprehensive review with which to assess the accuracy of the asserted average age
of relevant victims (i.e. only those ‘collateral victims) at 27 years28.
91.2 Whilst the Canadian database of the University of Sherbrooke does support the
Congolese average life expectancy age of 52.11 years as adopted by the DRC, my
review of the database does not provide clear support for USD 753.20 which is
asserted by DRC as being the country’s GDP29 per head for the year 201530.
91.3 Given that the Armed Activities relevant to this matter took place during the years
1998 to 2003, the DRC’s rationale for adopting the year 2015 for the GDP per head
data point is not beyond debate. While I understand that the DRC asserts 2015 on the
basis that this was the particular year’s level of (average) income that members of its
population would aspire to, however, in my opinion, this is not a robust basis on which
to assert losses of income that, continuing the averaging methodology adopted by the
DRC, may have commenced from as early as 1998 (some 17 years prior to the year
chosen by the DRC).
92. Based on the above brief synopsis of the DRC evidence in the record, together with my own
assessments thereof, I am of the opinion that my recommended individual flat rate
compensation amounts for consideration by the Court should not follow the flat rate amounts
claimed, but instead should be built on an alternative basis; one that has been already adopted,
implemented and paid out by a well-known, multi-billion dollar international mass claims
programme.
93. The United Nations Compensation Commission (the “UNCC” or “Commission”) was established
in 1991 as an agency of the UN Security Council31 with a mandate to accept, process, decide on
and pay compensation to successful claimants who suffered losses as a result of the military
hostlities and armed activities of Iraq’s 1990/91 invasion and occupation of Kuwait32. The
present report is not suitable for me to provide a detailed description of the Commission’s work
and decisions.33
94. The UNCC’s mandate included accepting claims from individuals and corporate entities, and
covering a wide scope of of losses; both pecuniary and non-pecuniary in nature34. Of most direct
relevance to the present matter under review, were the UNCC’s methodologies, decisions and
awards of compensation in respect of losses attriubuted to injury to persons.
28 DRC Memorial, para. 7.09.
29 Gross Domestic Product (or Produit Interne Brut [PIB] in French)
30 DRC Memorial, para. 7.09.
31 UN Security Council resolution 687 (1991): S/RES/687 (1991) 3 April 1991
32 For comprehensive details of the Commission’s structures and its work, including its awards and decisions,
see: www.uncc.ch
33 The author of this section, Geoffrey Senogles, worked on staff at the UNCC in Geneva between 2000 and
2003, and was engaged thereafter as an external independent consultant to continue advising Panels of
Commissioners on their valuation decisions.
34 In this way, there are similarities with the present matter before the Court.
46
95. To date, just short of USD 50 billion has been paid out by the United Nations Compensation Fund
to over 1.5 million successful claimants35.
96. The UNCC documented its decisions and made these freely available to all on the Commission’s
website. Of specific relevance to the DRC’s claim for non-pecuniary losses suffered by
individuals resulting from injury to persons, in late-1991 and early-1992 as part of establishing its
own valuation methodologies, the UNCC helpfully set out its definitions of seven separate
categories of mental pain and anguish, in the context of personal injury. These categories and
their definitions are set out in full in UNCC Governing Council Decision no. 3, as follows36:
“Mental pain and anguish”
Compensation will be provided for pecuniary losses (including losses of income and medical
expenses) resulting from mental pain and anguish. In addition, compensation will be provided for
non-pecuniary injuries resulting from such mental pain and anguish as follows:
(a) A spouse, child or parent of the individual suffered death;
(b) The individual suffered serious personal injury involving dismemberment, permanent or
temporary significant disfigurement, or permanent or temporary significant loss of use or
limitation of use of a body organ, member, function or system;
(c) The individual suffered a sexual assault or aggravated assault or torture;
(d) The individual witnessed the intentional infliction of events described in subparagraphs (a),
(b) or (c) on his or her spouse, child or parent;
(e) The individual was taken hostage or illegally detained for more than three days, or for a
shorter period in circumstances indicating an imminent threat to his or her life;
(f) On account of a manifestly well-founded fear for one's life or of being taken hostage or
illegally detained, the individual was forced to hide for more than three days; or
(g) The individual was deprived of all economic resources, such as to threaten seriously his or her
survival and that of his or her spouse, children or parents, in cases where assistance from his or
her Government or other sources has not been provided.”37
97. Of these seven categories, I regard the following two as being comparable in general to the
described claims associated with individuals’ deaths or injuries, and hence may be found
applicable as benchmarks for the Court’s findings in this matter.
“3. (a) A spouse, child or parent of the individual suffered death;
35 https://uncc.ch/summary-awards-and-current-status-payments showing a total of USD 49,964,258,680 has
been paid, up to and including the most recent quarterly payment made on 27 October 2020.
36 I set this out in full here, to provide context, since these definitions will be returned to in later sections of
this report.
37 https://uncc.ch/sites/default/files/attachments/documents/S-AC.26-DEC%2… UNCC
Governing Council decision no. 3. S/AC.26/1991/3 - 23 October 1991
47
(b) The individual suffered serious personal injury involving dismemberment, permanent or
temporary significant disfigurement, or permanent or temporary significant loss of use or
limitation of use of a body organ, member, function or system;...” 38
98. I note that the UNCC widened its definition of “serious personal injury” to include the following:
“2. For purposes of recovery before the Compensation Commission, "serious personal injury" also
includes instances of physical or mental injury arising from sexual assault, torture, aggravated
physical assault, hostage-taking or illegal detention for more than three days or being forced to
hide for more than three days on account of a manifestly well-founded fear for one's life or of
being taken hostage or illegally detained.”39
99. In January 1992, the UNCC’s Governing Council made its decision to ascribe monetary amounts
to each of the seven categories of mental pain and anguish, as defined by themselves three
months previously40. These compensation amounts were described in the decisions as ceilings
rather than fixed amounts, but in practice represented a form of tariff figures for application in
the Commission’s mass claims processing programme.
100. The individual compensation amounts ascribed by the UNCC in this way to its defined categories
A and B were as follows:
“CATEGORY A:
A spouse, child or parent of the individual suffered death.
- USD 15,000 ceiling per claimant;
- USD 30,000 ceiling per family unit.
CATEGORY B:
The individual suffered serious personal injury involving dismemberment, permanent or
temporary significant disfigurement, or permanent or temporary significant loss of use or
limitation of use of a body organ, member, function or system.
- USD 15,000 ceiling for dismemberment, permanent significant disfigurement, or permanent loss
of use or permanent limitation of use of a body organ, member, function or system;
- USD 5,000 ceiling for temporary significant disfigurement or temporary significant loss of use or
limitation of use of a body organ, member, function or system. “41
My recommendations to the Court
101. Using the abovementioned UNCC jurisprudence as a methodological basis and a benchmark, my
recommended flat rate compensation amounts in respect of “Human lives lost” are developed
in the following paragraphs.
102. I am of opinion that the UNCC findings on non-pecuniary loss mental pain and anguish are
applicable to and useful in formulating recommended flat rate individual compensation
38 Ibid
39 Ibid
40 For full details, see UNCC Governing Council decision no. 8 – dated 27 January 1992: S/AC.26/1992/8
https://uncc.ch/sites/default/files/attachments/S-AC.26-DEC%208%20%5B19…
41 Ibid
48
amounts in the armed activities context found in the present case.
103. By way of reminder, the DRC’s individual claim amount for deaths/injuries caused by deliberate
targetting of civilian populations is USD 34,000 applicable to actions taking place between 1998
and 2003.
104. A comparator individual compensation figure awarded by the UNCC, in its Governing Council
decison 8 category A, was USD 30,000 (decided in 1992).
105. The two following counter-balancing factors should be considered in any assessment of the
applicability of the UNCC award of USD 30,000 to the circumstances of the present case:
105.1 An argument could be made for this figure to be increased by some inflationary
adjustment to take into account the time value of money over the approximate 10-
year period between 1992 and the period of Armed Activities42; and.
105.2 As a counterbalance, it should be recalled that the USD 30,000 individual amount was
found to be appropriate in the context of deaths and injuries caused by military
activities in Kuwait, a country whose residents on average enjoy a level of GDP per
head far in excess of that of the DRC’s residents, as illustrated in the following chart,
created by the University of Sherbrooke global database, known as “Perspective
Monde”43:
42 As defined in the present case; mid-1998 to mid-2003
43 I note that the University of Sherbrooke data for GDP per head for Kuwait is not available for years prior to
1995. However, for general comparative purposes according to the Sherbrooke database, Kuwait’s GDP per
head between say, 1995 and 2002 stayed in a narrow range between USD 54,987 and USD 60,591 (at
purchasing power parity in current international US dollars). International Monetary Fund database figures
which are available for periods even prior to 1990, show that Kuwait’s GDP per head in 1990 was
approximately USD 24,500 (at purchasing power parity in current international US dollars): Source: IMF
database online at: https://www.imf.org/en/Publications/WEO/weo-database/2018/April/weorepo…?
c=443,&s=PPPPC,&sy=1988&ey=2018&ssm=0&scsm=1&scc=0&ssd=1&ssc=0&sic=0&sort=country&ds=
.&br=1
49
106. Based on my above analysis, and taking both of the counter-balancing factors into account, I am
of the opinion that USD 30,000 is a reasonable figure with a precedent for its use in analagous
circumstances and thus can be recommended for the Court’s consideration for application to
each individual death/injury found to have been caused by the deliberate targetting of civilian
victims.
107. Similarly, the DRC’s individual claim amount for deaths/injuries suffered by non-targetted
civilian populations or ‘collateral victims’ in 1998-2003 is USD 18,913; and is based on a
methodology of assuming an average victims’ future earning capacity, that averaged the
victims’ age at death, and that used the DRC’s asserted figure for GDP per head in 201544.
108. A comparator amount awarded by the UNCC, in its Governing Council decison 8 category B, was
USD 15,000 (decided in 1992). In my view, the same two counter-balancing adjustments could
potentially have been considered but these can reasonably seen to be neutral overall.
109. Based on my above analysis, and taking both counter-balancing issues into account, I am of the
opinion that USD 15,000 is a reasonable and reasoned figure to recommend for the Court’s
consideration for application to each individual death/injury for collateral victims.
110. In summary, my recommended amounts are shown alongside the claimed amounts in the table
below:
Claimed
amount
Recommended
amount
Deaths/injuries resulting from acts of violence
deliberately targeted at civilian populations
USD 34,000 USD 30,000
Deaths/injuries not resulting from violence targeted at
civilian populations but rather, as collateral victims
USD 18,913 USD 15,000
B) Injuries and mutilations – CLAIM: USD 54,464,000
111. The DRC asserts claims for injuries suffered in two categories, broadly similar to those
distinguishing circumstances in the previous claim line item, as follows:
111.1 Injuries resulting from acts of violence deliberately targeted at civilian populations45;
111.2 Injuries resulting from violence against civilian populations as collateral victims46.
112. Although both causes are without doubt harrowing, the more aggravated circumstances of an
alleged deliberate targetting of civilians, leads the DRC to attach a higher monetary amount on
the individual claim for each death/injury so caused, when compared to the analagous
individual amount in the (somewhat) less grevious circumstances where a civilian was killed or
injured even though not specifically targetted in the attack.
44 To reiterate, I have been unable to verify the asserted figure of USD 753.20 for DRC’s GDP per head in 2015
from the University of Sherbrooke’s database cited by DRC as its source, i.e. www.perspective.usherbrooke.ca
45 DRC Memorial, paras. 7.17-7.18.
46 DRC Memorial, para. 7.19.
50
113. The DRC’s claim computations are based on the following individual amounts:
113.1 Injuries resulting from acts of violence deliberately targeted at civilian populations:
- flat rates are based on Congolese court awards.
- individual claim amounts of USD 3,500 for a serious injury or USD150 for a minor
injury47
113.2 Injuries resulting from violence against civilian populations as collateral victims:
- a flat rate based on Congolese ordinary court awards.
- for minor injuries, an individual claim amount of USD 10048.
114. No supporting evidence is provided by the DRC for the USD 3,500 claimed amount that is stated
to be based on judgements of Congolese courts for serious injuries. The relevant court awards
are stated to range from USD 550 to USD 5,000, with the average being USD 3,50049. In the
absence of the court documents, I have no evidentiary basis on which to assess this claim figure.
I therefore will rely on the jurisprudence of the UNCC.
115. In this way, as quoted above, I rely upon the UNCC category B definition and its award amounts
in respect of personal injury, as follows:
“CATEGORY B:
The individual suffered serious personal injury involving dismemberment, permanent or
temporary significant disfigurement, or permanent or temporary significant loss of use or
limitation of use of a body organ, member, function or system.
- USD 15,000 ceiling for dismemberment, permanent significant disfigurement, or permanent
loss of use or permanent limitation of use of a body organ, member, function or system;
- USD 5,000 ceiling for temporary significant disfigurement or temporary significant loss of
use or limitation of use of a body organ, member, function or system. “50
116. In my opinion the DRC’s claimed amount of USD 3,500 for injuries can reasonably be
benchmarked against the lower of the above two grades of injury – with a UNCC award amount
of USD 5,000 - and hence it follows that the claimed amount of USD 3,500 can be recommended
as reasonable.
117. The other two (relatively) minor injury claim amounts are asserted by the DRC without evidence
that would have substantiated them as being derived from various Congolese court judgements
but, due to their nominal values, my opinion is they that can be recommended without any
need for adjustment.
47 DRC Memorial, para. 7.17.
48 DRC Memorial, para. 7.19.
49 DRC Memorial, paras. 7.17-7.18.
50 UNCC Governing Council decision no. 8 – dated 27 January 1992: S/AC.26/1992/8
51
My recommendations to the Court
In summary, my recommended amounts are shown alongside the claimed amounts in the
table below:
Claimed
amount
Recommended
amount
Injury resulting from acts of violence deliberately
targeted at civilian populations
Based on Congolese court awards:
- Serious injury USD 3,500 USD 3,500
- Minor injury USD 150 USD 150
Based on Congolese ordinary courts:
- Minor injury USD 100 USD 100
Injury not resulting from violence targeted at
civilian populations but as collateral victims
Based on Congolese court awards:
- Ituri: serious injury USD 3,500 USD 3,500
- Ituri: minor injury USD 150 USD 150
Based on Congolese ordinary courts:
- Eastern Congo, Ituri, Kisangani: minor USD 100 USD 100
C) Incidences of rape – CLAIM: USD 33,458,000
118. The DRC asserts claims for rape injuries suffered in two categories, as follows51:
118.1 “Simple” rape52;
118.2 Aggravated rape.
119. The DRC’s claim computations are based on the following individual amounts:
119.1 Injuries resulting from “simple” rape:
- flat rate of USD 12,600 based on Congolese court awards53.
119.2 Injuries resulting from aggravated rape:
- a flat rate of USD 23,200 based on Congolese court awards54.
120. Given the described circumstances of the incidences of rape, it is unsurprising and reasonable
that no documentary evidence is provided on an individual basis. In my review of the victim
identification forms, I note that victims indicated positively where appropriate.
121. I note however that in its responses to questions of the Court, the DRC provided contradictory
information to the amounts claimed and outlined above by stating that the Congolese court
51 DRC Memorial, paras. 7.08 and 7.22.
52 The term “simple” is the DRC’s own chosen terminology. I do not condone its use in relation to alleged rape.
53 DRC Memorial, para. 7.23.
54 DRC Memorial, para. 7.23.
52
awarded USD 5,000 in respect of rape injuries.
122. It is therefore the case that there is no supporting court evidence from the DRC for the USD
12,600 and USD 23,200 claimed amounts stated to be based on judgements of Congolese
courts. In the absence of the court documents, I have no evidentiary basis on which to assess
these claim figures. I therefore will rely on the jurisprudence of the UNCC.
123. In this way, as quoted above, I rely upon the UNCC category C definition and its awarded
amount in respect of sexual assualt or aggravated assault, as follows:
“CATEGORY C:
The individual suffered sexual assault or aggravated assault or torture.
USD 5,000 ceiling per incident.”55
124. Using the abovementioned UNCC jurisprudence as a methodological basis and a benchmark,
and noting that the DRC stated that its courst have also awarded this same figure in rape cases,
my recommended amount for rape is a flat rate of USD 5,000.
My recommendation to the Court
125. My recommended flat rate amount is shown alongside the claimed amounts in the table below:
Claimed
amount
Recommended
amount
Based on Congolese court awards:
- “Simple” rape USD 12,600 USD 5,000
- Aggravated rape USD 23,200 USD 5,000
D) Child soldiers – CLAIM: USD 30,000,000
126. The DRC claims compensation in respect of trauma of children being torn from their families
and their loss of education and life chances56.
127. The DRC’s claim is not based on evidence of loss, but rather is asserted for each individual
affected in an amount that the DRC deems reasonable, at USD 12,00057.
128. I note that in its responses to questions of the Court, did not provide any further information
regarding this claim item.58
129. Given the circumstances of each child soldier, in my opinion it is no unreasonable that no
documentary evidence is provided on an individual basis.
55 UNCC Governing Council decision no. 8 – dated 27 January 1992: S/AC.26/1992/8
56 DRC Memorial, paras. 7.26-7.28.
57 DRC Memorial, para. 7.27.
58 DRC Responses to questions of the Court, question 10.
53
130. It is the case that there is no supporting quantum evidence from the DRC for the USD 12,000
individual59. In the absence of the court documents, I have no evidentiary basis on which to
assess these claim figures and will therefore rely on the jurisprudence of the UNCC.
131. In this way, I rely upon UNCC category E definition and its awarded amount by way of an
analagous trauma caused by being detained and being in imminent threat to life, as follows:
“CATEGORY E:
The individual was taken hostage or illegally detained for more than three days, or for a
shorter period in circumstances indicating an imminent threat to his or her life.
USD 1,500 per claimant for being taken hostage or illegally detained for more than three
days, or for a shorter period in circumstances indicating an imminent threat to life;
USD 100 per day for each day detained in Iraq or Kuwait beyond three;
Ceiling of USD 10,000 per claimant.”60
132. Using the abovementioned UNCC jurisprudence as a basis and a benchmark, my recommended
amount for trauma caused to a child soldier is a flat rate of USD 10,000.
My recommendation to the Court
My recommended flat rate amount is shown alongside the claim amount below:
Claimed
amount
Recommended
amount
Claim amount deemed reasonable by the DRC:
- Child soldier USD 12,000 USD 10,000
E) Population flight and displacement – CLAIM: USD 186,853,800
133. The DRC claims compensation in respect of harm and trauma caused to individuals as a result of
being forced to flee from home due to armed hostilities61.
134. The DRC’s claim computations are based on flat rates applied in the following cases, based on
location and military units involved:
134.1 In Ituri, due to Ugandan breaches:
- a flat rate of USD 30062.
134.2 In Kisangani, plus Eastern Congo (outside Ituri):
- a flat rate of USD 10063.
135. No supporting evidence is provided by the DRC for these two flat rates claimed.
136. Given the described circumstances of the civilians’ flight into the forest and other surrounding
areas seeking physical safety, it is in my opinion unsurprising and quite reasonable that no
59 DRC Memorial, paras. 7.28 and 3.36.
60 UNCC Governing Council decision no. 8 – dated 27 January 1992: S/AC.26/1992/8
61 DRC Memorial, paras. 7.29-7.30.
62 DRC Memorial, para. 7.30.
63 DRC Memorial, para. 7.31.
54
documentary evidence has been provided by the DRC on an individual basis. In my review of
the original French language victim identification forms submitted, I note that individuals
indicated positively where appropriate (fuite dans le foret).
137. I note with interest that the DRC refers to UNCC compensation awards as benchmarks in respect
of civilians being forced to flee due to military hostilities64. By way of comparison, UNCC awards
for population flight and displacement started at a lump sum of USD 1,500, increasing by a daily
rate of USD 50, and with a ceiling set at USD 5,000 per individual. With these benchmarks in
mind, the flat rate amounts claimed by the DRC can be seen in some relevant context.
138. The DRC’s claim is not based on evidence of loss, not unreasonably in my opinion, but rather is
asserted for each individual affected in flat rate amounts that the DRC deems reasonable, at
USD 300 and USD 100.
138.1 My opinion is that the two flat rates claimed by the DRC for this category of loss are
seen to be not unreasonable at USD 300 and USD 100.
My recommendation to the Court
My recommended amounts are shown alongside the claimed amounts below:
Claimed
amount
Recommended
amount
Based on Congolese court awards:
- Ituri, due to Ugandan breaches: USD 300 USD 300
- Kisangani and Eastern Congo (outside Ituri): USD 100 USD 100
64 DRC Memorial, para. 3.24.
55
Summary of recommended amounts
139. In summary, for each category my recommended amounts are shown alongside the claimed
amounts in the table below:
Claimed
amount
Recommended
amount
A) Human lives lost
Deaths/injuries resulting from acts of violence
deliberately targeted at civilian populations
USD 34,000 USD 30,000
Deaths/injuries not resulting from violence targeted at
civilian populations but rather, as collateral victims USD 18,913 USD 15,000
B) Injuries and mutilations
Injury resulting from acts of violence deliberately
targeted at civilian populations
Based on Congolese court awards:
- Serious injury USD 3,500 USD 3,500
- Minor injury USD 150 USD 150
Based on Congolese ordinary courts:
- Minor injury USD 100 USD 100
Injury not resulting from violence targeted at civilian
populations but as collateral victims
Based on Congolese court awards:
- Ituri: serious injury USD 3,500 USD 3,500
- Ituri: minor injury USD 150 USD 150
Based on Congolese ordinary courts:
- Eastern Congo, Ituri, Kisangani: minor USD 100 USD 100
C) Incidences of rape
Based on Congolese court awards:
- “Simple” rape USD 12,600 USD 5,000
- Aggravated rape USD 23,200 USD 5,000
D) Child soldiers
Based on a figure deemed reasonable by DRC: USD 12,000 USD 10,000
E) Population flight and displacement
Based on figures deemed reasonable by DRC:
- Ituri USD 300 USD 300
- Eastern Congo and Kisangani USD 100 USD 100
56
SECTION: Property damage
140. In their Memorial65, the DRC presents claims for losses to property organised according to
the geographical location and the nature of properties relevant.
141. The two geographical areas adopted by the DRC, along with the component claims for each
area, can be summarised in the table below (with references to the DRC Memorial)66:
A) Property in Ituri – CLAIM: USD 41,524,61367
142. The DRC Memorial shows that the claim for Property Damage losses in Ituri fall into three
categories, as summarised below:
DRC Memorial
paras.
Claim amount
USD
Destruction of dwellings 7.35-7.37 12,956,200
Destruction of infrastructure 7.38-7.42 21,250,000
Looting 7.43 7,318,413
41,524,61368
Destruction of dwellings
65 DRC Memorial, paras. 7.33-7.53
66 In this table I show instances of apparent arithmetic inaccuracy with some figures in the DRC Memorial.
67 As restated to correct the arithmetic in the DRC Memorial.
68 As restated.
57
143. The claim for destruction of dwellings (in Ituri)69 is calculated in a systematic way, using a
methodology or formula which while being intuitive, by using a broad brush approach by
categorising the properties into (only) three grades and adopting a round-sum replacement
cost for each grade, almost inevitably leaves the resulting computations subject to
uncertainty and a lack of precision due to the absence of granular detail or evidence in
respect of each individual property.
144. At the outset of this section, I can make my view clear that given the prevailing context of
the Armed Activities having caused the civil disturbances in general, and specifically the
physical destruction to such significant numbers of properties70 – the vast majority of which
are designated by the DRC as having been “light”71 or basic structures mainly in rural areas -
it is understandable (and in my view, not unreasonable) for the damages claim in respect of
thousands of individual properties to have been formulated in such a way.
145. The following step-by-step description shows how the claim amount was derived:
Destruction of dwellings (Ituri)
DRC designates three grades of dwellings; basic, medium and luxury.
A replacement cost basis is used, with the following claim amount for each grade:
Basic USD 300 [para 7.35]
Medium USD 5,000
Luxury USD 10,000
The claim amount is built up based on the destruction of 8,693 dwellings in Ituri (para. 3.45):
[para 3.45c] [para 3.45c]
Basic 80% x 8,693 = 6,954
Medium 15% x 8,693 = 1,304
Luxury 5% x 8,693 = 435
To which the claimed replacement cost is applied, deriving the total claim amount:
Basic 6,954 x USD 300 = USD 2,086,200
Medium 1,304 x USD 5,000 = USD 6,520,000
Luxury 435 x USD 10,000 = USD 4,350,000
USD 12,956,200
146. According to its paragraph 3.45(c), the DRC Memorial cites analysis of the tabulated results
of its own investigations conducted in Ituri as the source and basis for the total number of
69 DRC Memorial, paras. 3.42-3.46
70 For example, as described in DRC Memorial, paras. 3.42-3.44
71 As officially translated into English from the original French term légère used by the DRC.
58
dwellings destroyed as being 8,69372.
147. In the same paragraph, the Memorial goes on to break down the total figure into the three
grades of dwellings using its 80%, 15% and 5% percentages shown above73. The basis for
these chosen percentages is not evidenced, but rather that it is stated that “... DRC considers
it reasonable to break the figure down into the following categories of destroyed buildings:
...”74
148. My review of the original language Annexe 1.3 reveals a different total number of properties
destoyed in Ituri (13,609) and also a different pattern for the split between their designated
grades, as follows75:
Basic 98% 13,384
Medium 1% 199
Luxury 1% (de minimis) 26
149. The replacement cost figures (USD 300, USD 5,000 and USD 10,000) are not evidenced and
are not explained in the DRC Memorial76.
150. My desk research into potential property replacement costs reveals no useful data. A UN
Habitat field report (2015) from a sustainable basic construction project in Eastern DRC
provided useful building ideas and recommendations, but no cost data77.
151. That said, in my opinion the claimed replacement costs asserted by the DRC are not
unreasonable in their amounts, particularly given that the overwhelming majority of the
properties are being valued for claim purposes at USD 300 each.
152. Accordingly, based on my analysis of the available evidence as described above, my
recommended amount can be calculated as follows:
Basic 13,384 x USD 300 = USD 4,015,200
Medium 199 x USD 5,000 = USD 995,000
Luxury 26 x USD 10,000 = USD 260,000
USD 5,270,200
My recommendation to the Court
My recommended amounts are shown alongside the claimed amounts below:
72 This figure is cited to the DRC Memorial, Annexe 1.3 (the memorial citation reads as follows in its footnote
316: “Result of using the software created by the DRC for the present proceedings (by entering “Ituri”
and “Destruction of property”).”
73 DRC Memorial, para. 7.35.
74 DRC Memorial, paras. 3.45 (c).
75 Source: DRC Memorial, Annexe 1.3, page 3, « Liste biens perdus et leurs fréquences ITURI.pdf », line items
no. 118 [habitation de luxe, 26], no. 119 [habitation légère, 13384], no. 120 [habitation moyenne, 199].
76 DRC Memorial, para. 7.35.
77 UN Habitat, Nairobi, website: https://unhabitat.org/sites/default/files/download-managerfiles/
Sustainable_Housing_Reconstruction_in_the_Eastern_Democratic_Republic_of_Congo.pdf
59
Claimed
amount
Recommended
amount
Destruction of dwellings
- Basic USD 2,086,200 USD 4,015,200
- Medium USD 6,520,000 USD 995,000
- Luxury USD 4,350,000 USD 260,000
USD 12,956,200 USD 5,270,200
Destruction of infrastructure
153. The claim for destruction of infrastructure is calculated in a similar way as for dwellings
(above), in that the DRC adopt an approach which does not provide granular details in
evidence for each of the various built structures claimed for78.
154. The infrastructure claim can be summarised as follows:
Schools79 200 x USD 75,000 = USD 15,000,000
Clinics80 50 x USD 75,000 = USD 3,750,000
Administrative81 50 x USD 50,000 = USD 2,500,000
USD 21,250,000
155. The figure for the number of schools destroyed can be verified to a report of the Secretary
General of the United Nations on the MONUC mission, dated 27 May 200382.
156. The figures for both clinics and adminsitrative buildings are DRC estimates. That they are
round sums, as in the case of dwellings (above), inevitably makes them subject to
uncertainties due to an absence of detail or evidence in respect of each individual property.
157. No evidence is provided in respect of any of the round sum values for any of the types of
infrastructure buildings. Due to a complete lack of any basis, rationale or evidence provided
by DRC in this regard, in my opinion the values claimed should be reduced by an evidentiary
discount factor of 25% by way of seeking to take account of the inherent uncertainty in the
way this claim has been put forward.
158. Accordingly, based on my analysis and opinion of the available evidence and assertions of
the DRC as described above, my recommended amounts can be calculated as follows:
Schools83 200 x USD 56,250 = USD 11,250,000
Clinics84 50 x USD 56,250 = USD 2,812,500
Administrative85 50 x USD 37,500 = USD 1,875,000
USD 15,937,500
78 DRC Memorial, paras. 3.45 (a) and Annexe 3.6, para. 10.
79 DRC Memorial, para. 7.39.
80 DRC Memorial, para. 7.40.
81 DRC Memorial, para. 7.41.
82 DRC Memorial, para. 7.41.
83 DRC Memorial, para. 7.39.
84 DRC Memorial, para. 7.40.
85 DRC Memorial, para. 7.41.
60
My recommendation to the Court
159. My recommended amounts are shown below, alongside the claimed amounts:
Claimed
amount
Recommended
amount
Destruction of infrastructure
- Schools USD 15,000,000 USD 11,250,000
- Clinics USD 3,750,000 USD 2,812,500
- Administrative USD 2,500,000 USD 1,875,000
USD 21,250,000 USD 15,937,500
Looting
160. DRC’s claim in respect of losses arising from the looting of property is asserted in the
amount of USD 7,318,41386. Although several harrowing events are described in the DRC
Memorial87 and supported by reports in its annexes, this amount claimed for pecuniary loss
is not well supported by evidence.
161. There is no breakdown of the claimed figure, even though the DRC Memorial states that the
total figure was derived from records generated by its own investigators visiting affected
regions of the country88.
162. Given this absence of clarity or breakdown of evidence, I have no practicable evidentiary
basis on which to assess the claim amount put forward. In these unfortunate circumstances
which creates unavoidable uncertainties as to the extent and the value of moveable assets
which was lost in the looting of properties, I am of the opinion that an evidentiary discount
factor of 50% should be applied in deriving a compensation award.
163. Applying this evidentiary discount factor of 50% would derive a recommended amount of
compensation of USD 3,659,206.
My recommendation to the Court
164. My recommended amount is shown below, alongside the claimed amount:
Claimed
amount
Recommended
amount
Looting USD 7,318,413 USD 3,659,206
86 DRC Memorial, para. 7.43.
87 DRC Memorial, para. 3.47 and, for instance, Annexe 1.06 (in its para. 73).
88 DRC Memorial, para. 7.43.
61
Summary of recommended amounts -
Property Damage (Ituri)
165. In summary, for each category my recommended amounts are shown alongside the claimed
amounts in the table below:
Claimed
amount
Recommended
amount
Destruction of dwellings
- Basic USD 2,086,200 USD 4,015,200
- Medium USD 6,520,000 USD 995,000
- Luxury USD 4,350,000 USD 260,000
USD 12,956,200 USD 5,270,200
Destruction of infrastructure
- Schools USD 15,000,000 USD 11,250,000
- Clinics USD 3,750,000 USD 2,812,500
- Administrative USD 2,500,000 USD 1,875,000
USD 21,250,000 USD 15,937,500
Looting USD 7,318,413 USD 3,659,206
Total - Property Damage (Ituri) USD 41,524,613 USD 24,866,906
B) Property in Kisangani and Ugandan occupied territory –
CLAIM: USD 192,457,357
166. The DRC Memorial shows that the claim for Property Damage losses in Kisangani fall into
three categories, as summarised below:
DRC Memorial
paras.
Claim amount
USD
Property in four named locations 7.46 25,628,075
Property of la Société Nationale
d'Electricité
7.47 97,412,090
Property of Congolese armed forces 7.48 69,417,192
192,457,357
Property losses in four named towns
167. The claim for property lost in the following four named locations is described in some detail
in the DRC Memorial, across various sections of the document89 and in several annexes.
89 DRC Memorial, for instance in paras. 4.22-4.26; 4.30-4.32; 4.48-4.54; 4.60 (list) and 7.45-7.46.
62
168. I can summarise the claimed amounts in the following table:
DRC Memorial
Annexes.
Claim amount
USD
Kisangani 4.03_F.pdf 17,323,998
Beni 2.04bis_F.pdf 5,526,527
Butembo 2.04ter_F.pdf 2,680,000
Gemena 2.04quater_F.pdf 97,550
25,628,075
169. I have reviewed the well organised lists of property losses produced from declarations by
individuals that are available for Beni, Butembo and Gemena and I have been able to carry
out manual additions deriving figures that closely approximate the figures presented above.
The figures supported by the lists in the DRC Annexes are as follows:
a. Beni USD 5,551,42790
b. Butembo USD 2,680,03091
c. Gemena USD 86,38092
170. The challenges faced in the process of DRC investigators being able to collate and detail
losses, given the prevailing circumstances, are not underestimated. The evidentiary basis is
commendable but is not, in my opinion, complete, fully detailed or supported by
documentation on the assets’ market values or historic costs – even though, I recognised
that setting such an evidentiary threshold may have been unachievable in the prevailing
circumstances.
171. However, to take into account these abovementioned evidentiary shortcomings, and hence
to recognise the risk of overstatement in the claim, I am of the view that an evidentiary
discount factor of 25% should be applied to the evidenced amounts when calculating my
recommended amounts (below) in respect of Beni, Butembo and Gemena.
172. In respect of Kisangani, the presentation and organisation of the available evidence by DRC
is less clear and less satisfactory. I have reviewed the annexe cited by the DRC93 and note
two deficiencies - it appears to be incomplete and provides no systematic breakdown of
component claim amounts building up to the total amount is provided. While the
document lists many instances of described losses, no support for figures can be clearly
seen. It is not unreasonable for the Court to have expected DRC to assume and achieve the
burden of collating and presenting a fully particularised breakdown of component claim
amounts from all of the various third-party source documents cited, which would have
90 DRC Memorial, Annexe 2.04bis_F.pdf, page 94
91 DRC Memorial, Annexe 2.04ter_F.pdf, page 32
92 DRC Memorial, Annexe 2.04quater_F.pdf, page 6
93 DRC Memorial, Annexe 4.03_F.pdf
63
allowed for a clearer and more robust claim to be assessed by the Court.
173. To take account of the evidentiary and presentational issues identified, and hence an
associated risk of overstatement in the claimed figures, I apply an evidentiary discount
factor of 40% in deriving my recommended amount in respect of Kisangani.
My recommendation to the Court
174. In summary, for each named location recommended amounts are shown alongside the
claimed amounts below:
Claimed
amount
Recommended
amount
Kisangani94 USD 17,323,998 USD 10,394,399
Beni95 USD 5,526,527 USD 4,163,570
Butembo96 USD 2,680,000 USD 2,010,022
Gemena97 USD 97,550 USD 64,785
USD 25,628,075 USD 16,632,776
Property of la Société Nationale d'Electricité SA (“SNEL”)98
175. The SNEL claim for losses is supported by a 17-page report dated 31 May 201699 prepared by
a nine-member committee of company employees100, which on 9 June 2016 was sent to the
Congolese President, the Prime Minister and other cabinet ministers.
176. On page 4 of their report, SNEL provides a table that breaks down seven categories of losses
claimed, across three Congolese provinces, as follows:
94 Recommended amount is: USD 17,323,998 (claimed) x 60%
95 Recommended amount is: USD 5,551,427 (evidenced) x 75%
96 Recommended amount is: USD 2,680,030 (evidenced) x 75%
97 Recommended amount is: USD 86,380 (evidenced) x 75%
98 DRC Memorial, para. 7.47.
99 DRC Memorial, Annexe 4.26 (available only in the original French language)
100 As identified by name and job title on page 15 of the SNEL report.
64
Province Claim amount
USD
North Equateur 49,207,607
North Kivu 2,428,608
Orientale 45,775,875
97,412,090
177. I have reviewed the subordinate tables within the SNEL report and note several of the
tables’ totals do not agree (i.e reconcile) with the summary table shown in the image above.
This is an unfortunate error and a matter of detail that the DRC can, in my opinion,
reasonably have been expected to rectify before submission of their claim.
178. Other observations on the SNEL report include the following:
a. The SNEL committee laid out a list of challenges they faced in fulfilling their mandate
of identifying, valuing and evidencing all relevant losses of their company – a list
which appears reasonable given my understanding of the prevailing circumstances
including the time elapsed between date of the Armed Activities in 1998-2003 and
the date of the SNEL report in 2016.
b. The valuation metholodogy adoped by SNEL is that of (new) replacement cost, even
though many assets destroyed or lost were identified as haveing been aged at vthe
time. Accordingly, in my opinion it would have been more appropriate for the DRC
and SNEL to present this claim using a ‘depreciated replacement cost’ approach.
65
c. Included within the claim figures are the cost of new replacement equipment along
with ancillary service costs to install.
d. No annexes containing underlying details or evidence appear to have been
referenced by, or attached to, the SNEL report.
e. Included within the list of asset loss categories is a claim for USD 6,543,953 for loss
of revenues by two of SNEL’s isolated hydroelectric centres located in Kisangani and
Gbadolite. I note a methodological overstatement in the calculations. SNEL have
claimed for an 8-year period of war (“1998-2005”) which goes beyond the findings
of the Court in its December 2005 judgement. Hence, this claim amount needs to be
reduced from eight to five years (before any other evidentiary factor is applied). The
resulting claim would be reduced from USD 6,543,953 to USD 4,089,970 (a reduction
of USD 2,453,983).
179. Thus, based on a review of the evidence provided for this claim element, I am of the opinion
that SNEL has prepared and submitted a report which seeks to systematically evaluate its
own losses. Whilst this is extremely useful, the report suffers from certain evidentiary
deficiencies in its efforts to substantiate SNEL’s claim. The report provides no detailed backup
calculations or underlying evidence supporting the various replacement costs or services
claimed. Also, the adoption of a ‘new for old’ replacement cost approach requires to be
adjusted for since, as SNEL acknowledge, much of the equipment destroyed was already old
and heavily used. In addition, the overstatement in the loss of revenue claim item must be
adjusted for, but leads to an inference that other claim components may also contain
overstatements.
180. In summary, to the adjusted claim figures an evidentiary discount factor of 40% should be
applied so as to recognise and take into account the abovementioned scope limitations,
methodological overstatements and other evidentiary gaps. Based on this, the resulting
recommended amount is calculated to be USD 56,974,865.
My recommendation to the Court
181. My recommended amount is shown below, alongside the claimed amount:
Claimed
amount
Recommended
amount
SNEL SA USD 97,412,090 USD 56,974,865
66
Property of the Congolese Military Forces101
182. The military forces’ claim for losses is supported by a two-page summary dated 31 August
2016102, stamped and signed by a General in the Congolese Army103.
183. The document lists 16 categories of arms, munitions, armoured vehicles and ships. The
majority of the line items are given a unit “value”104 and 9 of those unit costs can be agreed
to the second page of the document which lists unit costs (couts) for 27 military items. No
other evidence in support has been seen.
184. Notable high value items included in the list of lost military assets include:
a. Two ships (each one valued at over USD 21 million);
b. 400 tonnes of materiel and munitions (valued at USD 30,000/tonne, making a total
claim value of USD 12 million); and
c. 800 tonnes of munitions (valued at 10,000/tonne, making a total claim value of USD
8 million).
185. The above three asset categories amount to USD 68,350,000 which represents over 98% of
the total military assets’ claim.
186. Given the materiality of these three line items, I would have expected to see documentary
support that could have included:
a. Evidence in supporting for the events that caused the loss of each vessel, including the type,
age and identifying name of each vessel ;
b. Evidence supporting the vessels’ claimed unit value/cost of USD 21,375,000; and
c. Evidence for the unit value/cost of each tonne of munitions.
187. In the absence of further details it has not proved possible for me to independently verify
the claimed loss of these significant (and potentially individually identifiable) military assets’
or indeed, their unit values.
188. Taking the above analysis into account, I am of the opinion that signiificant evidentiary gaps
remain – gaps that the DRC should reasonably have foreseen and rectified in advance of
submitting their claim to the Court. I am therefore left with being required to use an
evidentiary discount factor to reflect the uncertainties and the potential for overstatement.
In this respect, I adopt an evidentiary discount factor of 40% which results in a
recommended amount of USD 41,650,315.
101 DRC Memorial, para. 7.48 and chapter 2.
102 DRC Memorial, Annexe 7.04 (available only in the original French language)
103 DRC Memorial, Annexe 7.04, page 2, General Damas Kabulo.
104 It is noted that the French term valeur is used rather than, for instance cout (cost).
67
My recommendation to the Court
189. My recommended amount is shown below, alongside the claimed amount:
Claimed
amount
Recommended
amount
Property of the Congolese Military Forces USD 69,417,192 USD 41,650,315
Summary of recommended amounts
190. In summary, for each category recommended amounts are shown alongside the claimed
amounts in the table below:
Claimed
amount
Recommended
amount
Property in four named locations USD 25,628,075 USD 16,632,776
Property of la Société Nationale d'Electricité USD 97,412,090 USD 56,974,865
Property of Congolese armed forces USD 69,417,192 USD 41,650,315
USD 192,457,357 USD 115,257,956
68
Appendix 3.1: Signature of Expert
Signature of expert
This report has been prepared in accordance with the terms of reference set out by the International
Court of Justice by Geoffrey Senogles on 19 December 2020:
Signed:__________________________
69
70
Report 4
Exploitation of Natural Resources
Dr. Michael Nest
(Montreal, 19th December 2020)
71
Table of Contents
List of Figures, Maps and Tables 72
1. Introduction 73
1.1 Selection of resources for consideration 74
2. Approach to estimating quantity and value 74
3. Quantity of resources produced and geographic distribution 75
3.1 Impact of conflict on production 76
3.1.1 Impact on opportunity for exploitation in Kisangani 76
3.2 Estimating resource quantities in Ugandan Area of Influence 78
3.2.1 Rationale for estimates of distribution 80
3.2 Distribution of resource production: Ituri v. non-Ituri 83
4. Resources Prices 85
5. Estimating the Exploitation of Value 92
5.1 Theft 93
5.2 Fees & Licences 95
5.3 Tax on sales, exports and other value 97
Appendix 1: Terms of Reference 101
Appendix 2: List of References 102
Appendix 3: Commodity Codes 107
Appendix 4: Reported taxes on natural resources 108
Appendix 5: Calculations of Value 111
A4.5.1 Gold 111
A4.5.2 Diamonds 117
A4.5.3 Coltan 120
A4.5.4 Tin 123
A4.5.5 Tungsten 125
A4.5.6 Timber 127
A4.5.7 Coffee 129
Appendix 6: Signature of Expert 131
72
List of Figures, Maps and Tables
Figures Page
4.1 Comparison of coltan prices 88
4.2 Comparison of sawn wood prices 90
4.3 Comparison of coffee prices 91
Maps
4.1 The DRC Party’s map of Ugandan occupation, 1993-2003 77
4.2 Diamondiferous areas in the DRC 81
Tables
4.1 Est. of quantity of resources produced, 1998-2003 78
4.2 Annual average resource prices, by year 86
4.3 Resource production in 2020 US dollars before value exploited 91
4.4 Est. value exploited by personnel: UAI, Ituri and non-Ituri 92
4.5 Est. of proxy taxes on theft and fees & licences, and tax on profits as
percentages
92
4.6 Est. proxy tax rate for theft only 94
4.7 Est. proxy tax rate for value of fees and licences 97
4.8 Tax range and adopted tax on value 98
4.9 Value of exploitation disaggregated by method, Ituri and non-Ituri, 2020 USD 100
73
1. Introduction
191. The ICJ provided terms of reference (TOR) to guide this report - see Annex 1.
192. In keeping with these TOR, this report focuses on the subset of quantity and value that was
“illegally exploited” in the Ugandan area of influence (UAI) in the territory of the Democratic
Republic of the Congo (DRC) between 1998 and 2003. Amounts for the entire DRC were
obtained only if needed to derive estimates of quantity and value of resources relevant to
the UAI, including in Ituri and outside Ituri (‘non-Ituri’).
193. The activity of exploiting value from resources was defined as falling into two categories:
193.1 In the UAI outside Ituri (in non-Ituri) when undertaken by UPDF personnel only.
This means any exploitation by, for example, personnel of the Mouvement de
Libération du Congo (MLC), is excluded from this report.
193.2 Within Ituri when undertaken by any and all armed forces and any affiliated
administrative personnel, including both UPDF and Congolese.
194. The period of “6 August 1998 to 2 June 2003” stated in the TOR was interpreted as
comprising 58 months including the entire month of August 1998, until the end-May 2003.
I.e., only the last five months of 1998 and only the first five months of 2003. The years of
1998 and 2003 in tables in this report represent only calculations for these five-month
periods and not the whole year.
195. The estimated total quantity of each resource produced is in Table 1. All quantities are in
kilograms, except diamonds which are in carats.
196. The estimated total value each resource before any exploitation by personnel is
$401,174,017 in 2020 US dollars105, of which $141,229,808 (35%) is in Ituri and $259,944,211
(65%) is in non-Ituri (see Table 3).
197. The total estimated value of exploitation activities by personnel in the UAI is $55,809,542,
which includes $38,986,151 (70% of value extracted) in Ituri, and $16,823,390 in non-Ituri
(30% of value extracted). These data are in Table 4.
198. Detailed tables showing calculations used to arrive at estimates for value for each resource
exploited in the UAI, including Ituri and non-Ituri, are at Annex 5: Calculations of value.
199. Numbers were not rounded: Data in this report appear precise because calculations were
used that produced precise amounts, e.g., percentages of whole figures or annual prices
to two decimal places. Instead of rounding numbers to the nearest thousand, which
would be the standard approach to avoid the perception of precision, the full numbers
are presented for transparency about the calculations. Rounding in the case of resources
valuable in small quantities, such as gold and diamonds, could also cause an unreasonable
loss or gain for one of the Parties of several million dollars.
105 All dollar amounts used in this report are United States dollars (USD).
74
1.1 Selection of resources for consideration
200. The TOR specified a requirement to identify “…the approximate quantity of natural
resources, such as gold, diamond, coltan and timber, unlawfully exploited during the
occupation by Ugandan armed forces …”. This report includes three additional resources -
tin (cassiterite), tungsten (wolframite) and coffee - for the following reasons:
200.1 Gold, diamonds, coltan and timber were identified as examples of resources (“such
as”). These four were therefore not considered to be an exclusive list.
200.2 Cassiterite (tin ore) extracted in DRC, is often found in the same ore body as
niobium-tantalite (coltan). A report on tin mining in eastern DRC noted “Generally,
in the mines of the region cassiterite, coltan and ferro-oxides coexist in the same
mineral and are separated manually, using pans and sifting” (Johnson and Tegera
2005: 49). There is no good reason to include coltan in this report but exclude tin.
200.3 Scholarship on production and value of natural resources during 1998-2003,
including publications at Annex 2: References consulted for this report, typically
include analyses of what is known as the ‘3Ts’: tin, tantalite and tungsten, as well
as gold and diamonds. Excluding tin and tungsten given the attention paid to these
resources would be an error due to intense interest in these minerals and their
connection to conflict in DRC.
200.4 In regard to coffee, UNPE (2001a; 2001b; 2002a; 2002b) and MONUC (2004)
specifically include coffee in their reports, including some limited data on exports
from DRC into Uganda. Neglecting coffee, in my view, would be an error.
201. This report estimates that limited value was exploited from tin and tungsten. However,
given public interest in these resources they have been included in order to flag their
relative insignificance as sources of value exploited by personnel in either Ituri or non-Ituri.
2. Approach to estimating quantity and value
202. Estimating the quantity and value of selected resources involved eight basic steps:
202.1 Identifying the distribution of resources within UAI.
202.2 Estimating the distribution of each resource between Ituri and non-Ituri in the
form of a percentage.
202.3 Estimating the quantity of resources produced.
202.4 Estimating the percentage of value extracted by different methods of exploitation.
202.5 Estimating an appropriate price per unit (kilogram or carat) for each resource.
202.6 Estimating value exploited from these resources by personnel.
202.7 Adjusting the value of exploitation into 2020 USD to reflect current prices.
202.8 Estimating value of each resources exploited in Ituri and non-Ituri.
75
203. Exploiting value was understood in three ways:
203.1 Theft
203.2 Value extracted in the form of fees and licences
203.3 Value extracted in the form of taxes on trade and exports
204. Specific instances of theft, taxation or payments of fees and licences are described in case
file and other documents. However, complete information necessary to estimate exploited
value during such incidents was missing in virtually all cases, e.g., quantity, value, location or
approximate date. (Date has a significant bearing on price as prices fluctuated between
August 1998 and June 2003, in some cases substantially, such as for coltan).
205. Furthermore, it was assumed that every such incident that ever occurred within the UAI
during the time period was not documented or made available in the case file documents.
Therefore, the totality of quantity and value exploited was assumed to extend beyond
documented instances of theft, taxation or payments for fees and licences.
206. Although some data about exploitation of value were available in case file documents, much
of the required data was absent. The incompleteness of data meant other sources of
information had to be relied on to inform estimates about resource distribution and
quantities, including maps of deposits, anecdotal descriptions of resource distribution from
field observations in THE DRC, or production data had to combined from several sources.
207. Specific sources of information used to estimate quantity and value are mentioned in notes
to tables in Annex 5: Calculations of Value.
3. Quantity of resources produced and geographic distribution
208. There were several challenges in estimating quantities of resources and their distribution
between the UAI and outside the UAI, as well as Ituri and non-Ituri:
208.1 Available data are incomplete and it was often unclear what portion of production
came from UAI.
208.2 Production in the early to mid-1990s was significantly affected by tumultuous
conditions, so it cannot be considered as a baseline.
208.3 Production during 1998-2003 was interrupted by conflict, general breakdown of
infrastructure and disruption of commerce, complicating efforts to estimate
probable production based on pre-conflict data.
208.4 Significant production of all seven resources occurred using artisanal small-scale
means, meaning production and trade was often not recorded.
208.5 There was significant smuggling of all seven resources which meant this portion of
trade was missing from trade data and challenging to quantify.
76
3.1 Impact of conflict on production
209. The impact of conflict on production within the UAI from 1998 to 2003 varied for each
resource:
Gold 210. Industrial production collapsed in non-Government-held areas, where the
bulk of THE DRC’s gold is found. However, artisanal production probably
increased, propelled by armed groups’ financial imperative to obtain
revenue. (HRW, 2005; Johnson and Tegera, 2005; Mthembu-Salter 2015a;
OHCHR, 2003)
Diamonds 211. Industrial extraction has historically been confined to Government-held
areas, not relevant to this report. In non-Government-held areas, artisanal
production of diamonds probably had a modest increase due to armed
groups’ financial imperative to obtain revenue. (DIAR, 2005; Dietrich, 2002;
GAO, 2002; Goreux, 2001)
Coltan 212. Industrial production in non-Government-held areas had ceased prior to
conflict in 1998-2003. The conflict coincided with greatly increased global
demand for tantalite, causing a boom in prices and artisanal production.
(International Alert, 2010; IPIS, 2002; Johnson and Tegera, 2002; Le Billon and
Hocquard, 2007; Nest, 2011; Raeymaekers, 2002; Redmond, 2001).
Tin 213. Industrial production had ceased but artisanal production increased during
the conflict due to cassiterite being frequently found in the same ore-bearing
body as tantalite-niobium (miners for one often ended up extracting the
other as well) and due to a surge in global demand for tin from 2003.
Cassiterite production probably increased gradually during the period, then
exponentially at the end. (Global Witness, 2005; International Alert, 2010;
Johnson and Tegera, 2002; Nest, 2011)
Tungsten 214. Industrial production had ceased. Artisanal production in non-Governmentheld
areas probably increased significantly but from a low base.
(International Alert, 2010)
Timber 215. Conflict caused major commercial production to crash throughout THE DRC.
Timber harvesting was restricted to timber stocks relatively close to Kinshasa
or to the eastern border in non-Government-held areas, with cutting almost
exclusively by artisanal means. Informal production in non-Government-held
areas probably increased due to reduced controls over commercial
concessions. (Baker et al, 2003; Chatham House, 2020; Counsell 2006;
Megevand 2013; Umunay 2011).
Coffee 216. Commercial production throughout DRDC crashed. Artisanal production
continued in non-Government-held areas, but probably declined overall.
(ICO, 2020; Kamungele, 2013; Wilkins, 2019).
3.1.1 Impact on opportunities for exploitation in Kisangani
217. Conflict affected the relationship between Uganda and Rwanda and their respective THE
DRC allies in the form of contested influence over Kisangani, a key strategic city with two
airfields in Orientale Province. The Porter Commission (2002: 121-122) noted: “Evidence
shows that Kisangani, though not a diamond producing area in itself, was the basis of
77
collection and distribution. It was also Victoria [company’s] base”. The Victoria Company
was a company connected to UPDF General Kazini that traded in “diamonds, gold and
coffee” (Porter Commission 2002: 82).
218. In June 2000, after several violent clashes between Rwandan and Ugandan troops over 1999
and 2000, and “Although it maintained control of most diamond producing regions, the
UPDF was defeated by the [Rwandan Patriotic Army] in the city itself” (ICG 2000b: 8-9).
UPDF personnel remained stationed 12kms away at a nearby airport.
219. Loss of influence in Kisangani meant UPDF personnel had reduced opportunities to exploit
value from resources traded there (Baker et al 2003; HRW 2001; ICG 2000a; ICG 2000b;
Porter Commission). Ugandan loss of influence is factored into calculations about quantity
and value exploited from gold and diamonds, as is explained in notes in Annex 5:
Calculations of Value.
220. Map 1 was submitted by the DRC Party (2016: §9.27) to show the geographic distribution of
UPDF personnel (the shaded area). The distribution of UPDF personnel was not
comprehensive, consistent or constant within the shaded area. I.e., UPDF personnel were
not everywhere all the time, with the same level of strength, within the shaded area. The
map is, nevertheless, a useful tool to match to reports of resource distribution.
Map 4.1: The DRC Party’s map of Ugandan occupation, 1998-2003
Source: DRC Memorial 2016: §9.27
78
3.2 Estimating resource quantities in Ugandan Area of Influence
221. Information about resources produced in UAI falls into two categories: (a) formal production
that is recorded, and (b) production that escapes most records, typically because it is
produced in the informal sector and is smuggled out of the country. This report takes both
types of production and trade into account.
222. Where national data for resources were not available or appeared too unreliable, export
and/or import data for countries trading in the DRC resources were used to estimate
probable production within the UAI. For example, there are ComTrade data of countries
reporting imports of tin, coltan, tungsten, coffee and sawn wood (timber) from the DRC
(data recorded by importers was used because DRC export data in ComTrade is so
incomplete). Such data were used as a ‘proxy’ for DRC production.
223. Given what is known about the location of each resource (derived from a mix of case file and
other documents), an estimate was then made to understand what percentage of these
imported resources probably came from UAI. These calculations are explained in notes to
tables in Annex 5.
224. Table 1 shows estimated quantities of resources for the entire the DRC (where these could
be obtained) and for the UAI, including both Ituri and non-Ituri.
Table 4.1: Est. of quantity of resources produced, 1998-2003
DRC UAI Ituri Non-Ituri
Quantity Quantity
% of
DRC
Quantity
% of
UAI
Quantity
% of
UAI
Gold, kgs 39,896 22,106 55.4 9,949 45 12,158 55
Diamonds, carats 96,372,668 4,260,627 4.4 213,031 5 4,047,596 95
Coltan, kgs ? 84,082 ? 4,204 5 79,878 95
Tin, kgs ? 890,428 ? 44,521 5 845,907 95
Tungsten, kgs ? 330,825 ? 16,541 5 314,284 95
Timber, kgs ? 89,369,380 ? 44,684,690 50 44,684,690 50
Coffee, kgs ? 43,779,341 ? 13,133,802 30 30,645,539 70
225. For both gold and diamonds there are national production data that could be used to make
estimates for UAI when combined with information about the location of resources. These
data were the most reliable available for these resources.
226. For other resources, ComTrade import and export data was used to estimate production in
UAI. The commodity codes used for ComTrade database searches are in Annex 3.
227. Unrecorded and smuggled production was significant for all seven resources:
227.1 OHCHR (2003: 363) states “A large part of the gold produced in Ituri was exported
through Uganda, then re-exported as if it had been produced domestically – a
similar model to that used for diamond exports”. [Emphasis added]
227.2 The Porter Commission notes:
227.2.1 “There is no doubt in our minds that diamonds are being smuggled and
falsely declared as sourced in Uganda” (p.114).
79
227.2.2 UPDF officer, General Kazini, was “an active supporter in the Democratic
Republic of the Congo of Victoria, an organization engaged in smuggling
diamonds through Uganda” (p.122).
227.2.3 “On the 31st of December 1998 General Kazini messaged Major Kagezi,
saying that his soldiers and detach commanders were writing chits for gold
mining and smuggling and instructing him to stop this immediately” (p.19),
indicating some UPDF personnel’s involvement in smuggling.
227.2.4 One witness from Arum told the Commission that “…10 to 20 trucks a day
were transporting timber from Congo to Uganda without paying taxes. The
trucks use 20 feeder roads and join the main road after the customs post.
While the Commission thought that the number of daily trucks was
exaggerated, there was no doubt that smuggling of timber at that point was
actually taking place … There was clear evidence from a Congolese who lives
near the border in the Democratic Republic of Congo of daily smuggling of
timber over the border to Uganda” (p.153).
227.3 Baker et al (2003) writing about ‘conflict timber’ in the DRC argues:
227.3.1 “…despite strong global demand for high-quality tropical timber, as a
commodity it: requires more skills and more expensive equipment to
harvest; and has a much lower weight-to-value ratio, and is often both
difficult to transport (especially if transportation infrastructure is
dilapidated, as in the DRC) and to conceal. Smuggling is thus not an
attractive option; instead, collusive arrangements have to be organized with
government officials (customs officers, border guards, etc.) to avoid formal
controls. Collusive arrangements often depend on bribes, and these drive up
transaction costs associated with the operation” (p.14).
227.3.2 Thus Baker et al argue that it is not smuggling in the form of timber
“unseen” by officials that was occurring between the DRC and neighbouring
countries. Rather, the “smuggling” involved collusion with officials,
especially border officials, to ensure timber is trafficked into neighbouring
countries such as Uganda without being recorded. Officials extract a portion
of the timber’s value in the form of bribes.
228 Estimates of informal production and trade were taken from publications that contained
estimates, such as PAC (2004) for diamonds, Mthembu-Salter (2015) for gold and Umunay
(2011) for timber.
229 For other resources, the gap between Uganda’s production of a resource and the quantity of
its exports was identified. When there were more exports than production it was assumed
this ‘surplus’ was originated in the DRC. The DRC was assumed to be the source, rather than
neighbouring countries, due to reports of Ugandan involvement in certain resources and the
small probability of a resource originating from countries other than the DRC (or Uganda
itself, depending on the resource). The Porter Commission noted resources originating in the
DRC being imported by other countries via Uganda: “…most of the resources flown or driven
out of the Democratic Republic of Congo appear to have transited Uganda, rather than to
have been exported to Uganda…” (p.85).
80
230 Where potential non-DRC sources for Ugandan exports exist and the origin of the resource
can be clearly confined to a subset of countries - such as coltan, tin and tungsten from
Rwanda or Burundi - these quantities were subtracted from Ugandan export data (as is
explained in notes to Annex 5: Calculations of Value).
3.2.1 Rationale for estimates of distribution
231. Below is an explanation of the sources of data, other estimates for informal unrecorded
production or trade, and assumptions made to estimate the proportion of resources in: the
Government-held area v. non-Government-held area (where this was necessary); within
latter what proportion was in the UAI v. the non-UAI (where this was necessary); and within
the UAI what proportion was in Ituri v. non-Ituri:
Gold 232. National data on formal production of gold is taken from USGS Minerals
Yearbook: Gold.
233. Data on informal artisanal production is based on Mthembu-Salter’s estimate
(2015a) of 10-15 metric tonnes nationally (not just UAI), which also
references an International Peace Information Service (IPIS) artisanal
estimate for Orientale only of 16.5mt, as well as an estimate of annual trade
in artisanal gold through Kisangani and Bunia of 5.40mt. IPIS (2013: 13)
estimated gold production for Ituri of about 2,000 kgs per year (presumably
based on research around 2012-2013, so almost a decade after 2003). IPIS’s
figure is around 14-20% of Mthembu-Salter’s national estimate, which is in
keeping with the latter in terms of geographic distribution of production
across the DRC.
234. Mthembu-Salter’s estimate of national artisanal production of approximately
15mt in the mid-2010s was reduced by two-thirds to reflect the probable
situation in 1998-2003, i.e., 5,000 kg annual national artisanal production
(artisanal production grew significantly after 2003). The lower end of HRW’s
(2005: 55) estimate of 1,000-3,000 kgs of gold leaving the “Mongbwalu area”
each month in 2004, is in keeping with this estimate of 5,000 kg per year.
235. For both formal and informal production, 80% of national production was
assumed to come from the non-government-held area (both UAI and non-
UAI) due Orientale’s known history as the DRC’s major gold-producing area
(DRC 2016; HRW 2003; HRW 2005; ICG 2000a; ICG 2000b; 2004; International
Alert 2010; IPIS 2003; OHCHR 2003; Spittaels and Hilgert 2010). 75% of nongovernment-
held formal and artisanal production was then estimated to be
in UAI to June 2000 and then 70% from July 2000 (from when Ugandan
influence in Kisangani was diminished).
236. Estimates for informal production for Uganda (see Annex 5, Table A4.5.1.3)
were based on Hinton (2012: 7). Informal production, along with formal
production, was taken into account in calculating what proportion of
Uganda’s gold exports are likely to have originated domestically compared to
from the DRC. Hinton’s estimate of informal production was 1,210 kg for
2008. The quantity of informal production was assumed to be lower for
1998-2003 because, as Hinton notes, there was an expansion in the informal
81
gold sector in Uganda in the mid-2000s. The estimate for informal production
was therefore reduced to 1,000 kgs per year for the 1998-2003 period.
Diamonds 237. Estimated industrial and artisanal production is taken from PAC (2004). These
data include industrial diamonds, such as produced by the parastatal
company MIBA in government-held territory, which are not relevant to
artisanal production in UAI.
238. Production in Equateur and Orientale was based on PAC (2004) using
Dietrich’s (2002: 2) estimate of 10% of production coming from these two
provinces. (With the exception of Lubero and Beni districts in northern North
Kivu for which there are few, if any, reports of diamond production 1998-
2003, the rest of UAI was in Equateur and Orientale). Dietrich’s estimate was
then reduced to 9% (i.e., a reduction of 10%) to allow for a margin of error in
his original estimate.
239. Of combined formal and artisanal production in Equateur and Orientale, 70%
was estimated to be in UAI to June 2000 (when Uganda shared influence in
Kisangani) based on reports of diamond mining and trading, and maps of
deposits (see, for example, Map 2). UAI’s share was reduced from 70% to
35% from July 2000. Kisangani’s significance was in the opportunities it
created for exploitation of value through levying fees, licences or taxes, or
demands for exclusive sales agreements on traders which forced them to
accept prices below free market rates (GAO 2002; Goreux 2001; ICG 2000a;
ICG 2000b).
Map 4.2: Diamondiferous areas in the DRC
Source: Dietrich, Monnaie Forte, 2002, p.5
Coltan,
Tin and
Tungsten
240. Estimates for informal production - the only method of production relevant
in the UAI 1998-2003 - for these three resources all followed the same steps.
First, DRC domestic production (USGS) was subtracted from ComTrade data
of imports from trading partners to identify any gap in unrecorded
production. Because known production of coltan, tin and tungsten in the
DRC 1998-2003 was overwhelmingly in non-government-held territory, 95%
82
of production was assumed to be from this area. Because there are few
reports of these three resources being produced in UAI (most was in North
and South Kivu or Maniema), only 5% of non-government-held production
was assumed to be from UAI (i.e., the overwhelming majority was in the
Rwandan area of influence, within the non-Government-held area).
241. Second, Ugandan domestic production (taken from USGS) was subtracted
from declared imports by Ugandan trading partners to identify any surplus
exports that are likely to have originated in the DRC.
242. Third, imports from countries in the region that do not produce any of these
minerals were then identified. A portion of imports was estimated to have
originated in the DRC and passed through Uganda (this percentage is noted
in Tables in Annex 5; it varies according to country). Intra-African trade was
excluded other than trade specifically with the DRC.
Timber 243. Production is of two types: that for which there was reporting (based on
ComTrade reports of exports from importers of DRC timber) and that for
which there is no data - informal production and trade (based on Umunay
2011). Data for formal production is, in fact, available from the International
Tropical Timber Organization (ITTO), but these data are apparently sourced
from DRC central government officials. Due to the conflict 1998-2003, such
officials would not have known either production or export levels relevant to
the UAI. This report therefore uses ComTrade import data instead.
244. Because the distribution of harvestable and transportable timber 1998-2003
was predominantly in the non-Government-held area, it was assumed 80% of
reported (formal) timber exports came from here. It was then assumed that
50% of timber in this area came from the UAI (based on harvestable forests,
proximity to the Ugandan border, and a road network that remained open
facilitating exports during this period).
245. Umunay estimated in 2011 (8 years after 2003) that informal sawn wood
exported from the DRC to Uganda, Kenya and Rwanda totalled around
70,000,000 kgs (100,000m3) annually. This figure, presumably based on 2010-
11 activity, is too high for 1998-2003 due to war and degraded infrastructure.
Furthermore, informal DRC exports to Kenya may have originated in UAI
creating opportunities for personnel there to extract value, but they may also
have passed through Rwanda.
246. This report assumes 60% of Umunay’s estimate of informal (unrecorded)
timber passed through Uganda (42m kgs), based on timber harvesting areas
in proximity to Uganda and the number of border crossings that could
facilitate informal trade compared to Rwanda. This report then took then
took 20% of 42m kgs to reflect probable levels of informal exports in 1998-
2003. That is, this report estimates informal timber production in UAI was
8,400,000 kgs per year (12% of Umunay’s total original estimate from 2011).
Coffee 247. Production was estimated from ComTrade reports of imports of DRC coffee.
248. Because the distribution of harvestable and transportable coffee 1998-2003
was predominantly in the non-Government-held area, it was assumed 80% of
83
reported coffee exports on ComTrade came from UAI. It was then assumed
that 50% of this coffee came from the non-Government-held area (based on
history of producing areas, proximity to the Ugandan border, and a road
network that remained open during this period).
249. Informal smuggled quantities of coffee entering Uganda from DRC were
estimated by subtracting both (a) Ugandan exportable production (taken
from International Coffee Organization) and (b) declared Ugandan imports of
coffee from the DRC, Kenya, Rwanda and Burundi (taken from ComTrade),
from recorded imports of coffee by Uganda’s trading partners (taken from
ComTrade). Positive amounts indicate left over, un-exported, production.
Negative amounts (which occurred in 1999 and 2003) mean Uganda
exported more than it produced raising the question as to the origin of this
coffee. As green coffee beans can be stored for one year, the years where
exports are surplus to production were ‘discounted’ by an amount equal to
50% of the previous full year’s production surplus, to allow for some coffee
being stored for 12 months before export the following year.
250. ComTrade data were used over ICO data for exports (from the DRC, Uganda
and other countries) because they were deemed more reliable as linked to a
specific trading partner with a specific value per year for that partner. ICO
data were, however, used to estimate exportable production for Uganda -
but not for the DRC because of the uncertainty around whether any
authority in the DRC 1998-2003 was capable of collecting accurate data on
coffee production. ICO data are also used for price comparison as shown in
Fig.3.
3.3 Distribution of resource production: Ituri v. non-Ituri
251. Distribution of the selected resources within UAI across Ituri and non-Ituri varies greatly.
Below is a summary of information and sources used to make the estimated distributions in
Table 1:
Gold 252. In UAI outside Ituri, gold is produced around Beni in northern North Kivu
(International Alert 2010: 18); around Durba and Watsa in Haut-Uélé,
including Kilo-Moto mine (DRC 2016: 119; HRW 2003: 12, 15; OHCHR 2003:
358; UNPE 2001a: §57, §59;); and around Bondo in Equateur (UNPE 2001a:
§59). Butembo (North Kivu) is a centre for trading gold (HRW 2005: 55).
253. In Ituri there is extensive gold production in the area lying northwest of Bunia
through to Mongbwalu and Kilo-Moto in Haut-Uélé (HRW 2005: 24, 34; HRW
2003: 23; Johnson and Tegera 2007: 75; MONUC 2004, 8; UNPE 2001a;
International Alert 2010: 20). Bunia is a center for trading gold (OECD 2015a)
as is Ariwara (HRW 2005: 104; MONUC 2004: 38).
254. Around 45% of gold production in UAI probably came from Ituri, and around
55% from non-Ituri.
Diamonds 255. In UAI outside Ituri, diamonds are reportedly produced in northern North
Kivu, Tshopo and Haut-Uélé as follows: in northern North Kivu, including near
Butembo and Lubero (Raeymaekers 2002: 21; International Alert 2010: 18);
in or in proximity to Bafsawende, Banalia, Basoko, Buta, Kisangani, Opala and
84
Isangi in Tshopo (Raeymaekers 2002: 13, 17; DRC 2016: 125); and around
Isiro, Dungu and Watsa in Haut-Uélé (Raeymaekers 2002: 13). Kisangani was
a major trading centre for diamonds and Isiro (Haut-Uélé) a minor trading
centre (DIAR 2005: 10).
256. Ituri has some diamond production northwest of Bunia (HRW 2003: 12; DRC
2016: 125), and Ariwara is a centre for trading diamonds (MONUC 2004: 38).
257. Around 5% of diamond production in UAI probably came from Ituri, and
around 95% from non-Ituri.
Coltan 258. In the UAI outside Ituri in northern North Kivu, tantalite-niobium is produced
in Beni-Lubero territories (HRW 2001; Porter Commission 2002: 182; OHCHR
2003: 355), and in Orientale (UNPE 2002a: §108). Beni and Butembo in
northern North Kivu are major trading centers for coltan (Johnson and Tegera
2005: 27; Raeymaekers 2002: 21).
259. HRW (2003: 12) has a reference to coltan production in Ituri but provides no
detail. IPIS (2013: 13) states “There is hardly any recorded tin, tantalum or
tungsten mining in Ituri. The artisanal mining sector revolves almost entirely
around the exploitation of gold”. “Hardly any” suggests there may have been
very minor quantities produced, but no further information is provided. The
IPIS Interactive Minerals Map106 indicates coltan mines in northern North Kivu
very close to, or even on, the border with Ituri from 2009. Thus, it is possible
there were some pre-2009 coltan mines in Ituri, but given there was no
specific information these were disregarded for the purposes of this report.
260. Probably all tantalite-niobium mined in UAI came from outside Ituri, but
some trade occurred through Ituri. For the purposes of this report, and taking
likely transit trade through Ituri into account, 95% of coltan was allocated to
non-Ituri and 5% to Ituri.
Tin 261. In UAI outside Ituri, there is reported cassiterite production in Beni-Lubero
territories in North Kivu (International Alert 2010: 18; Garrett 2008: 12).
UNPE (2001a: §53) reports exports of cassiterite into Uganda through
Mpondwe and ‘Bundbujyo’ (this may a reference to the Bujerere crossing in
Bundibugyo District, Uganda), and Garrett (2008: 35) reports exports from
the DRC into Uganda via Ishasha and Bunagana border crossings in North
Kivu, neither one part of UAI (Johnson and Tegera 2007: 26). It is unclear
where cassiterite passing through these crossings originates.
262. There are no reports of cassiterite production in Ituri, although it may have
been mined in conjunction with coltan from some deposits. Probably 100% of
cassiterite production was outside Ituri, but some trade occurred through
Ituri. For the purposes of this report, and taking likely transit trade through
Ituri into account, 95% of cassiterite was allocated to non-Ituri and 5% to
Ituri.
106 https://ipisresearch.be/mapping/webmapping/drcongo/v5/#-
1.4045201543864891/28.801971434596567/5.774660859437679/2/1/1.9.20,2.157bpc
85
Tungsten 263. In the UAI outside Ituri, there is a report of wolframite deposits in Lubero
northwest of Butembo (Spittaels and Hilgert 2013: 10). It is not entirely clear
if these were mined for the whole 1998-2003 period, but it is clear that
significant quantities of wolframite were exported from Uganda (in excess of
Ugandan production). It is therefore probable there were operational
wolframite mines within the UAI.
264. There are no reports of wolframite production in Ituri, but some trade in this
resource occurred through Ituri. For the purposes of this report, and taking
likely transit trade through Ituri into account, 95% of coltan was allocated to
non-Ituri and 5% to Ituri.
Timber 265. In UAI outside Ituri, timber is harvested around Tshopo (Orientale), parts of
Equateur, and northern North Kivu (Baker et al 2003: 22, 38, 65-66; Chatham
House 2020; Counsell 2006: 8-9; Megevand 2013: 30; UNPE 2001a). There
are reports of exports into Uganda across Orientale’s and North Kivu’s land
borders, including timber processed at Mangina near Beni, North Kivu (Baker
et al 2003: 57, 66-67; UNPE 2001b: §48). Given distance and transport
networks, it is unlikely any timber from Equateur, and only timber from
eastern parts of Orientale, went to Uganda.
266. Ituri produces timber and the town of Ariwara is a timber trading centre
(Baker et al 2003: 51; Chatham House 2020; HRW 2003: 12; MONUC 2004: 8;
Umanay 2011; UNPE 2002a: §116).
267. Around 50% of timber production was probably in Ituri and 50% in non-Ituri.
Coffee 268. In UAI outside Ituri, coffee is produced in North Kivu, Orientale and Equateur,
with reports of export into Uganda (Porter Commission 2002: 18; UNPE
2001a: §102; Wilkins 2009: 5). Given distance and transport networks, it is
unlikely much coffee from Equateur made its way to Uganda.
269. In Ituri there is some coffee production (MONUC 2004: 8; Wilkins 2009: 5).
270. Around 30% per cent of coffee in UAI probably came from Ituri, and around
70% from outside Ituri in eastern Orientale and northern North Kivu.
4. Resources Prices
271. Estimating the value of resources before exploitation by personnel involved three steps:
271.1 Identifying base annual average prices for 1998-2003 (either an international price
or a price specifically identified as relevant to the DRC, such as ComTrade data for
imports from the DRC).
271.2 Discounting base prices by an appropriate amount to reflect probable prices
relevant for producers, traders and exporters in UAI. This report calls this the
‘adopted price’.
271.3 Adjusting adopted prices into 2020 USD by ‘inflating’ them using a standard rate.
86
272. The value of resources change year-on-year due to price fluctuations. For example, gold in
2003 was about 30% more expensive than in 1999, and coltan had peak prices from
November 2000 to February 2001 that were ten times prices in 1998. Thus, rather than take
an average price for the entire 58 month period, an average annual price for each year was
adopted to obtain a more accurate figure (a monthly price would be even more accurate,
but these are impossible to ascertain for the full 1998-2003 from ComTrade data).
273. Estimates of value were based on prices at likely points of opportunity for exploitation,
including personnel’s contact with producers, small traders, large traders, and exporters.
This means that value is not based on global prices for a commodity, nor simply on the price
that producer may have received. Value is based on an average price from the multiple
points of contact along the supply chain prior to export.
274. Table 2 shows the price used for each commodity and year, including the base price, the
price estimated to be relevant within the DRC and the ‘inflation’ factor to estimate value in
2020 USD.
Table 4.2: Annual average resource prices, by year*
* Price is per kilogram except for diamonds, which is per carat
1998 1999 2000 2001 2002 2003
Gold base price 9,455.20 8,956.22 8,973.26 8,714.13 9,956.43 11,680.99
Adopted price (35% less) 6,145.88 5,821.54 5,832.62 5,664.18 6,471.68 7,592.64
Diamond base price 18.59 12.55 14.34 18.79 19.33 27.43
Adopted price (35% less) 12.09 8.16 9.32 12.21 12.56 17.83
Niobium-Tantalite
base price
12.98 47.90 114.62 86.73 47.24 14.11
Adopted price (35% less) 8.44 31.14 74.50 55.07 30.71 9.17
Cassiterite base price 3.27 2.31 2.82 3.12 3.10 6.35
Adopted price (35% less) 2.12 1.50 1.83 2.03 2.02 4.12
Wolframite base price 2.48 2.00 3.49 3.34 2.87 3.66
Adopted price (35% less) 1.61 1.30 2.27 2.17 1.86 2.38
Timber base price 0.67 0.67 0.52 0.62 0.52 0.64
Adopted price (35% less) 0.44 0.44 0.35 0.40 0.34 0.42
Coffee base price 2.04 1.71 1.42 1.18 1.04 1.06
Adopted price (35% less) 1.33 1.11 0.92 0.77 0.68 0.69
Inflator to est. 2020 USD
(adopted for all)
x 1.60 x 1.56 x 1.51 x 1.47 x 1.45 x 1.41
275. Resource tables in Annex 5: Calculations of Value note the source of prices and any
discounting to reflect probable prices within the DRC. However, below is an explanation of
the estimates used and comparisons with other prices where sufficient data were available
to do this.
Gold 276. Estimates are based on annual average international prices provided by the
World Gold Council online database
(https://www.gold.org/goldhub/data/gold-prices), which were ‘discounted’
35% to reflect probable prices within the DRC at points of opportunity for
exploitation.
87
277. Price comparison: HRW (2005), interviewing the leader of the Nationalist
and Integrationist Front (FNI) in October 2003 (after the period relevant to
this report), reported him calculating he would “make about $50,000 from
the sale of five kilograms of gold” (p.55). This amount is only 20% less than
world gold price for that month ($12,181 pkg), but the miners and traders
upstream from the FNI leader would have received prices within a range
around the 35% discounted gold used in this report.
Diamonds 278. Estimates are based on the average export price per carat for artisanal
production only (only artisanal production is relevant to UAI) provided in
PAC (2004: 2-3). The price adopted for estimates was then ‘discounted’ 35%
to reflect probable prices within the DRC across the range of points of
opportunity for exploitation, i.e., production, small traders, larger traders,
and exporters. Prices vary for gem diamonds (which are expensive)
compared to industrial diamonds (which are much cheaper). PAC’s data do
not differentiate between each kind of diamond, so the estimated price
obtained is an average across gem and industrial stones. Because it is not
possible to know the portion of diamonds from UAI that was gem or
industrial, using an average was the best approach.
279. Price comparison: There are very few price observations around value per
carat. However, Johnson and Tegera (2005: 97), presumably doing their
research around 2004 (after the 1998-2003 period), note “Emaxon buys
MIBA diamonds for only $13.40 per carat, less than the price paid to
artisanal miners” [emphasis added]. Indeed, this report’s price for 2003 (the
year closest to their research) is $17.83, so more than MIBA’s price in
keeping with Johnson and Tegera’s observation.
Coltan 280. Estimates are based on an average of all price observations107 for each year
(1998-2003) available from UN ComTrade’s (https://comtrade.un.org)
records for niobium-tantalite imports and exports involving East and Central
African producers. This price was then discounted by 35% to better reflect
probable prices at points of opportunities for exploitation in the DRC.
281. Price comparison: Johnson and Tegera (2005: 37) quote a mine manager
stating “In 2000 and 2001 business was good and we managed to sell a kilo
of coltan for up to $150”. Given prices on the international spot market
peaked in December 2000 at almost $600/kg, such prices at the mine ‘gate’
are likely, as is this report’s annual adopted average price for 2000 of
$74.50 within the DRC (prices soared as the year progressed). They also
note prices in August 2003 (after the period of interest) of “$14.4/kg for
30% grade coltan and $16.8 for 35% grade” (p.31). These are higher than
this report’s adopted 2003 average price of $9.15/kg, but given the latter is
an average and the price continued to fall that year it is probably
reasonable.
107 An average of price observations (i.e., separate trades in coltan by different countries within a single year)
was used rather simply the average price (total value divided by total quantity) for each year. This is because
the latter can be grossly influenced by a very large quantity involving a single trading partner, when what is
needed is information about the range of prices that can then be used to estimate a probable average price.
88
282. Redmond (2001: 11-12) writing in May 2001 when prices were falling after
the price spike in late 2000/early 2001, notes a DRC trader who bought
coltan from miners “at $12 per kilogram”. Redmond also publishes a list of
prices from a buyer in Kigali, Rwanda, based on tantalite concentrate (prices
were converted from pounds in the original into kilogram): 10% Ta at
$44/kg; 16% Ta at $110/kg; 18% Ta at $132/kg; and 20% Ta at $165/kg. The
adopted price used in this report for 2001 of $55.07 falls into the lower end
of this category (ore around 10-16% concentrate).
283. IPIS (2002: 11) documents two consignments of tantalite ore sold in
December 2000 by RCD-Goma’s trading monopoly, SOMIGL, that had prices
of $52.2/kg and $51.5/kg. These prices are lower than adopted for this
report, but the consignments were very large (each 30mt) which might have
meant a lower price per kilogram and I also do not know the percentage of
concentrate which is critical in determining price. The same report notes a
consignment with 22.5% concentrate from August 2001 priced at $8.50/kg.
This appears very price, notwithstanding the decline in prices throughout
2001. It cannot be reconciled with this report’s adopted prices, unless it was
deliberately undervalued for reasons of evading import duties.
284. Johnson and Tegera (2002: 9) refer to “trading posts” buying coltan in late
2001 with 40% tantalite concentrate at $26/kg. This is less than this report’s
adopted annual average for 2001 of $55.07/kg but not far off 2002’s price
of $30.71. As Johnson and Tegera’s reference is to the price paid by trading
posts and not exporters’ buying or sales prices, this report’s adopted prices
are probably reasonable.
285. UNPE (2002a: §109) interviewed in Kampala in March 2002 a co-owner of
coltan trading firm La Conmet, who said “their purchase price for coltan
with a 30 per cent tantalum content was $10 per kilogram and sold it on at
a price of $17 per kilogram”. This price is lower than the adopted price used
in this report for 2002, which is almost double. Fig. 1 compares the adopted
price with price observations mentioned.
Fig. 4.1: Comparison of coltan prices
286. On balance, given that miners received lower prices than traders (in
keeping with both Redmond’s and UNPE’s observations); given that prices
77
150
16
0.00 52 26 17
50.00
100.00
150.00
200.00
1998 1999 2000 2001 2002 2003
USD per kilogram
Adopted price Redmond (2001)
Johnson and Tegera (2005) IPIS (2002)
UNPE (2002)
89
in Rwanda would be higher than inside the DRC (relevant to Redmond’s
report); given that anecdotal reports of prices described here fall not too far
from either side of this report’s adopted prices (although there is a lack of
price observations for comparison 1998-1999), the adopted prices are
reasonable.
Tin 287. Estimates were calculated using the same method as for coltan.
288. Price comparison: Garrett (2008: 13) writing about North and South Kivu,
notes export grade cassiterite with a content of around 65% on the world
market fetched “…USD3,800 [per tonne] at the start of the new
millennium”, or $3.80/kg - “new millennium” was assumed to mean 2000-
2001. This is about double this report’s adopted price of $1.83 for 2000 and
not quite double that for 2001 ($2.03/kg), i.e., this report’s prices are on the
low side in comparison.
289. Johnson and Tegera (2005: 53) note “at the height of the boom” (2003-
2004) traders in Walikale (North Kivu) paid $2.50 per kilogram. This is
substantially less than the adopted price for 2003 ($4.12), but =it is for
traders purchasing from miners - a price always lower than anything further
along the value chain.
290. Notwithstanding being different to these two observations, this report’s
adopted prices are likely to be reasonable for the average prices received
by miners, small and larger traders, and exporters for these years.
Tungsten 291. Estimates were calculated using the same method as for coltan.
292. Price comparison: No reports of prices from the relevant period for
production and trade in the DRC were available for comparison with this
report’s adopted prices.
Timber 293. Price estimates are based on International Tropical Timber Organization’s
database for export prices per unit (m3) of non-coniferous tropical sawn
wood at https://www.itto.int/biennal_review/?mode=searchdata. ITTO
prices are in cubic metres and were converted into USD per kg (1m3 = 700
kg), then discounted 35%.
294. Price comparison: Baker et al (2003: 68) estimates timber prices (which
were converted from m3 into per kg) for each of the relevant years as
follows: 1998 (0.50/kg), 1999 (0.47/kg), 2000 (0.58/kg), 2001 (0.58/kg),
2002 (0.29/kg), and 2003 (0.36/kg). Baker et al base their estimate on
market rates for DRC-origin timber in Burundi for 1998 to 2001, and in
Rwanda for 2002 to 2003, and then reducing them by half.
295. Baker et al’s estimates are 6%-67% higher than this report’s adopted prices
for 1998-2001, but 14% lower for 2002-2003. This report’s prices for 1998-
1998 and 2002-2003 are reasonable given they fall within Baker et al’s
overall range for the period.
296. The markedly lower estimates used in this report for 2000 and 2001 are a
bit concerning but, given the focus on Uganda as the key market (not
Burundi), and given that “Congo timber is cheaper in [this] market because
90
it is usually cut by chain saws, which are not allowed in Uganda” (Porter
Commission 2002: 55) they are probably within the range of reasonable.
297. Fig. 2 is a graph comparing price observations from six sources: this report’s
adopted price (based on ITTO), Baker et al’s (2003) prices, the Food &
Agricultural Organisation’s (FAO) sawn wood price, ComTrade’s price
observations for Ugandan imports of sawn wood from the DRC, ComTrade’s
price observations for Ugandan exports of sawn wood to the world, and
Djiré (2003).
Fig. 4.2: Comparison of sawn wood prices
298. Fig. 2 shows that this report’s adopted price is at the low end of price
observations for 1998 and 2001, around the middle of the observations for
1999, 2000 and 2003, and the highest of two observations for 2002. On
balance, there is confidence the adopted prices are reasonable.
Coffee 299. Estimates are based on an average of all price observations for each year
(1998-2003) available from UN ComTrade’s (https://comtrade.un.org)
records for imports of coffee from the DRC. This price was then discounted
by 35% to better reflect probable prices at range of points of opportunities
for exploitation in the DRC, and is this report’s adopted price.
300. The DRC produces two types of coffee: milds (more expensive) and robusta
(cheaper). It is not possible to tell from ComTrade data the proportion of
milds and robusta for each year. ComTrade was, nevertheless, used instead
of ICO ‘Prices to Growers’ for two reasons. First, the latter had a DRC milds
price observation only for 1998. This meant an average with robusta could
not established 1999-2003 (an average would have been an obvious price
to adopt if it could be calculated). Second, ICO’s ‘Prices to Growers’ is for
producers - not others further along the value chain, such as traders and
exporters, who are highly relevant to this report because they constitute
opportunities for personnel to extract value.
301. Price comparison: This report’s adopted prices were compared with the
International Coffee Organization’s (ICO) ‘Prices to Growers’ Historical Data
on the Global Coffee Trade, for DRC and Uganda robusta 1998-2003, the
91
single observation for DRC mild (1998), and Uganda milds 1998-2003
(http://www.ico.org/new_historical.asp). Fig. 3 is a graph of these prices.
Fig. 4.3: Comparison of coffee prices
302. Fig. 3 shows the adopted price is consistently higher than Ugandan robusta
1998-2003, and about the same as Uganda milds 1999-2003. The adopted
price is significantly lower than DRC robusta for 1999 - a difference that
cannot be reconciled - but higher than DRC robusta 2001-2003 (which
should be expected given the latter is for producers only).
303. On balance, notwithstanding the anomaly of 1999, and given the absence of
prices for DRC milds 1999-2003, the adopted price is probably reasonable.
304. Table 3 shows the estimated total value of production for each resource during the entire
August 1998 to end-May 2003 period, in 2020 USD, including distribution between Ituri and
non-Ituri, after applying the prices listed in Table 2.
Table 4.3: Resource production in 2020 US dollars before value exploited
Total UAI Ituri Non-Ituri
USD % of UAI USD % of UAI
Gold 201,817,503 90,817,876 45 110,999,627 55
Diamonds 70,531,967 3,526,598 5 67,005,369 95
Coltan 4,385,250 219,263 5 4,165,988 95
Tin 3,009,245 150,462 5 2,858,783 95
Tungsten 959,380 47,969 5 911,411 95
Timber 51,632,189 25,816,095 50 25,816,095 50
Coffee 68,838,483 20,651,545 30 48,186,938 70
Total $ 401,174,017 $ 141,229,808 $ 259,944,211
92
5. Estimating the Exploitation of Value
305. Table 4 shows the total estimated value extracted by personnel, and disaggregation for this
amount by resource for Ituri and non-Ituri.
Table 4.4: Est. value exploited by personnel: UAI, Ituri and non-Ituri
Ituri Non-Ituri Total UAI
2020 USD % share 2020 USD % share 2020 USD % share
Gold 33,012,298.1 84.7 9,834,566.9 58.5 42,846,865.0 76.8
Diamonds 1,013,897.0 2.6 5,025,402.6 29.9 6,039,299.7 10.8
Coltan 63,038.0 0.2 312,449.1 1.9 375,487.0 0.7
Tin 43,257.9 0.1 214,408.7 1.3 257,666.6 0.5
Tungsten 13,791.1 0.0 68,355.8 0.4 82,146.9 0.1
Timber 2,793,301.4 7.2 645,402.4 3.8 3,438,703.8 6.2
Coffee 2,046,568.1 5.2 722,804.1 4.3 2,769,372.2 5.0
Total 38,986,151.6 100 16,823,389.6 100.1 55,809,541.2 100.1
306. The estimates in Table 4 were calculated using assumptions about the methods of exploiting
value. This report categorises the extraction of value from resources into three different
methods:
306.1 Theft
306.2 Fees and licences, including permission to extract, trade or export a resource
306.3 Tax(es) on the value of sales or exports
307. UNPE (2002b: §47) notes this variety of methods using its own phraseology in a comment
about the state of the economy in the UAI: “Excessive taxes, revenue siphoning, seizure of local
resources, forced requisitioning of assets and deepening control over general trade by foreign
and local military, with or without the collusion of commercial operators, have paralysed local
economies”.
308. This section outlines evidence that each method occurred in Ituri and non-Ituri, and explains
how the extraction of value was calculated.
309. Table 5 shows the estimated rates of extracting value used in this report for (a) theft -
expressed as a proxy tax); (b) fees and licences - expressed as a proxy tax; and (c) a tax on
value, such as sales or exports. Explanations for the rates are in the table.
Table 4.5: Est. of proxy taxes on theft and fees & licences, and tax on profits as percentages
A. Proxy Tax:
Theft
B. Proxy Tax:
Fees and Licences
C. Tax on Value:
Sales and Exports
Total Tax Rate
(A+B+C)
Ituri Non-Ituri Ituri Non-Ituri Ituri Non-Ituri Ituri Non-Ituri
Gold 5.0 2.0 5.0 2.0 28.0 5.0 38.0 9.0
Diamonds 5.0 0.5 5.0 2.0 20.0 5.0 30.0 7.5
Coltan 5.0 0.5 5.0 2.0 20.0 5.0 30.0 7.5
Tin 5.0 0.5 5.0 2.0 20.0 5.0 30.0 7.5
Tungsten 5.0 0.5 5.0 2.0 20.0 5.0 30.0 8.0
Timber 2.0 0.5 1.0 1.0 8.0 1.0 11.0 2.5
Coffee 1.0 0.0 1.0 0.5 8.0 1.0 10.0 2.0
93
5.1 Theft
310. In the UAI outside Ituri, there is evidence Congolese personnel stole resources. These events
are outside the TOR of this report.
311. In regard to UPDF personnel in the UAI outside Ituri, there are some reported cases of theft:
a. UNPE (2002a) describes a network of “high-ranking UPDF officers, private businessmen
and selected rebel leaders/administrators” (§98) that generates revenue from,
amongst other things “theft” (§100). It is assumed some of the theft involved natural
resources.
b. OHCHR (2003: 355-356) reports that UPDF soldiers “requisitioned gold” from the
OKIMO plant at Durba, near Watsa in Haut-Uélé, Orientale.
c. The DRC Party (2016, 114) states “On 23 May 2001, OKIMO’s management again
reported that illegal artisanal miners, overseen by Ugandan soldiers and the RCD-ML
co-ordinator, were occupying the Durba mine [Haut-Uélé] and company
infrastructure”.
d. UNPE (2001a; §34) reports that in August 1998 “General Kazini’s soldiers absconded
with stockpiles of timber belonging to the logging company Amex-bois, located in
Bagboka”. Bagboka is the airport approximately 11km east of Kisangani, Orientale,
where UPDF was stationed. (The Porter Commission contests this example, noting
shortly after the event Amex-bois continued to export timber to Uganda so clearly still
had timber. However, the fact Amex-bois may have had all, or most, of its timber
stolen does not necessarily mean it could not subsequently return to business.)
e. Baker et al (2003: 28) reports that during the relevant period “a sawmill in Butembo
was similarly looted of stocked lumber for export to either Uganda or Rwanda.” It is not
clear if the theft was by Congolese forces or UPDF personnel, but it is possible some
UPDF personnel were involved given they had a significant presence in Butembo,
northern North Kivu.
f. A Congolese NGO, Le Société Civile Grand Nord, based in North Kivu, reported in June
2001 that “UPDF troops commit acts of looting, theft, rape, killings, massacres and
arson in the villages. [Including] … complicity in trafficking in raw materials, fraudulent
dealings in coffee, timber, papaine [papaya], etc. In all cases the UPDF military operate
in collusion with Congolese rebel troops”108
g. Porter Commission (2002: 197) notes “Looting, about which General Kazini clearly
knew as he sent a radio message about it”, adding “This commission is unable to
exclude the possibility that individual soldiers of the UPDF were involved [in looting], or
that they were supported by senior officers.
312. In Ituri, based on information in the case file and other documents it is probable there was a
pattern of theft of resources by some UPDF personnel, as well as Congolese forces, similar to
108 See DRC (2016), Memorial Annex E Vol. 2: 19.
94
what occurred outside Ituri. There are some specific reports of some UPDF personnel
arranging, or engaging in, theft:
a. UNPE (2001a: §44) reports UPDF’s General Kazini appointed Adele Lotsove to the
position of chief administrator of Ituri to “facilitate looting activities” of natural
resources in that area.
b. UNPE (2002a: §116) reports UPDF personnel Colonel Peter Karim and Colonel Otafiire
helping to organise the raiding of tree plantations “…in the areas of Mahagi and Djugu
[both Ituri] along the north-eastern border with Uganda”. UNPE states this amounts to
“illegal logging and fraudulent evacuation of wood”, i.e., theft.
313. It is likely there was theft of resources in both Ituri and non-Ituri, although as a method of
extracting value it was probably less common than levying fees and licences or taxing profits.
314. It is necessary to make an estimate for theft notwithstanding inconsistent and incomplete
information about it or the high likelihood that theft extended beyond the examples mentioned
above. A proxy tax rate for theft is estimated in Table 6 (data are the same as those in Table 5):
Table 4.6: Est. proxy tax rate for theft
Resource Ituri (%) Non-Ituri (%)
Gold 5.0 2.0
Diamonds 5.0 0.5
Coltan 5.0 0.5
Tin 5.0 0.5
Tungsten 5.0 0.5
Timber 2.0 0.5
Coffee 1.0 0.0
315. Below is the rationale for these estimates:
Gold 316. There is a specific reference to theft of gold, including by UPDF personnel
(OHCHR 2003: 355-356). There is abundant gold in the UAI, there are many
reports armed forces personnel were deliberately located in mining areas in
order to extract value from gold, and there was an established network of mines
that were being exploited during the period. All these factors created
opportunities for theft.
317. Within Ituri all armed forces are likely to have stolen limited quantities of gold
from producers and traders. Outside Ituri, it is probable some UPDF personnel
engaged in limited theft of gold. However, in both areas the evidence suggests
theft was a minor method of extracting value.
Diamonds 318. There are general references to theft of natural resources by armed forces
(UNPE 2002a: §100; UNPE 2001a: §44), including in areas where diamonds are
found and including Ituri. Given the preparedness of armed forces’ personnel to
steal gold, it should be assumed these personnel were also prepared to steal
diamonds. However, diamonds are easy to hide and therefore harder to steal,
and if one cannot identify a rough diamond it can be difficult to know what to
steal. Outside Ituri, some UPDF personnel’s theft of diamonds is likely to have
been limited due to fewer opportunities and their concentration in gold95
producing areas, not diamond areas. In Ituri and non-Ituri the evidence suggests
theft was a minor method of extracting value.
Coltan,
Tin,
Tungsten
319. Given reported theft of other minerals, it is likely there was also theft of coltan,
tin and tungsten in Ituri by armed forces exploiting transit traffic and the export
trade to Uganda. At least one Ugandan border post’s records shows such trade
(UNPE 2001a: §102).
320. Outside Ituri, there was confirmed production of coltan and tin in northern
North Kivu. Deposits in Orientale were probably also exploited during the
relevant period. Given reports of theft of other minerals, it is reasonable to
assume some UPDF personnel stole minor quantities of these resources.
Timber 321. There are specific references to theft of timber (Baker et al 2003: 28; DRC 2016:
19), including by UPDF personnel (UNPE 2001a; §34; UNPE 2002a: §116). Timber
is found across the UAI, including Ituri, but it is bulky making it difficult to
transport. Even after 2003 timber production has remained small-scale (Baker et
al 2003, Counsell 2006, Megevand 2013, Umunay 2011), meaning that with the
exception of commercially accumulated stockpiles at the start of the 1998-2003
period, there probably only small quantities available in any one location,
limiting quantities available for theft.
Coffee 322. There are references to theft of natural resources, including coffee, and
including in coffee-growing areas in Ituri and non-Ituri. There is a specific
reference from a Congolese NGO in North Kivu (DRC 2016: 19). However, Coffee
is bulky and not very valuable in small quantities. During 1998-2003, coffee was
produced by smallholders meaning there were fewer points of accumulation of
large quantities. There were few incentives to steal coffee without an easy way
of exporting it outside DRC or reselling it locally. Within Ituri all armed forces
probably stole limited quantities of coffee. Outside Ituri, any theft of coffee by
UPDF personnel was probably minor.
5.2 Fees & Licences
323. The table in Annex 4: Reported taxes on natural resources lists information about the level of
fees, licences and taxes from the case file and other documents. This information informs this
section and the next (5.3. Tax on value of sales and exports).
324. Given that many reported fees, licences and taxes are from outside the UAI, or within the UAI
outside Ituri but only involving Congolese personnel, their usefulness for this report is to
understand the likely range of taxes that were levied at various points in the production, trade
and export value chain.
325. Observations about Annex 4:
325.1 Some time periods are outside the relevant period (Aug 1998 to end-May 2003).
325.2 Many references for are areas outside the UAI.
325.3 Most of the tax collecting entities reported do not appear to directly involve UPDF
personnel
96
325.4 Tax rates for coltan and cassiterite appear to be broadly similar, but rates for gold and
diamonds are quite different, i.e., there does not appear to have been a standard
‘resource tax’
326. Assumptions based on Annex 4:
326.1 It was assumed that the references to ‘minerals’ meant coltan (tantalite-niobium) and
cassiterite due to what is known about location.
326.2 UNPE (2001b: §44) states “The high combined taxes imposed by the RCD-Goma rebel
group and RPA ultimately resulted in diamonds mined in this area being redirected to
Kampala, where lower tax rates prevail”. This passage suggests taxes within the UAI, at
least for diamonds, were probably lower than those reported outside the UAI.
326.3 Taxes originally set by the RCD were the baseline on which the various subsequent RCD
factions would have based their own tax rates, i.e., RCD-Goma, RCD-Kisangani, RCDML,
RCD-National, and FLC (RCD-ML’s temporary union with MLC).
327. Case file and other documents include examples where producers, traders and exporters of
resources were charged fees and licences:
327.1 UNPE (2002a: §101) describes a network of “high-ranking UPDF officers, private
businessmen and selected rebel leaders/administrators” (§98) that established
“…authority in major urban and financial centres, such as Bunia [Ituri], Beni [North
Kivu] and Butembo (North Kivu), where [the network] use the rebel administration as a
public sector façade to generate revenue, specifically to collect taxes under various
pretexts, including licensing fees for commercial operators, import and export duties
and taxes on specific productions”.
327.2 UNPE (2002a: §108), discussing coltan, states “Armed groups frequently identified with
militias under the command of UPDF officers manage sites in remote locations where
diggers pay a daily fee to exploit an area”.
327.3 The Porter Commission (2002: 109) reports that around September 1999 Professor
Wamba “appointed a Commission of soldiers to charge artisanal miners at Kilo-Moto
about $15 worth of gold to go into the mine, and that the proceeds from that were
about two to three hundred grams a month”. That is, miners were charged an
‘entrance fee’ of $15 in the form of gold to mine. It is unclear whether the “soldiers”
were UPDF personnel or RCD-ML forces.
327.4 ICG (2003: 5) reports RCD-ML leaders Mbusa Nyamwisi and Tibasiima Ateenyi ousted
Jean-Pierre Bemba (leader, MLC) from Bunia in November 2001, after disagreements
over several issues, including refusing to “accept being deprived by Bemba and [UPDF
General] Kazini of the U.S.$100,000 levied on Congolese traders at the border posts of
Kasindi and Mahagi [both North Kivu]”.
328. There are also examples of RCD-Goma charging fees and licences, a practice that was likely to
be continued by all RCD factions outside of areas controlled by RCD-Goma (i.e. in the UAI):
328.1 Licences “for trading in agricultural products increased fourfold” from September 2000
to March 2002 (UNPE 2002a: §89).
97
328.2 Circa 2000, 22% of profits from coltan mined in North Kivu and South Kivu, and
collected by RCD-Goma, were spent on “licences and fees” (Le Billon and Hocquard
2007: 90).
328.3 RCD-Goma from Nov 2000 to April 2001 demanded a $40,000 annual fee from
organisations and individuals wishing to export coltan from the DRC (Congo European
Network, see DRC Memorial Annex E, Vol. 2: 15).
329. These examples are of Congolese organisations, not UPDF personnel. They are relevant to this
report because of the high probability the methods described were also used in Ituri.
330. In the case file and other documents, different values for fees and licences were given for
resources, in different locations, at different times, collected by different organisations and
individuals, and demanded of different organisations and individuals. UNPE (2002a: §105)
notes local commercial operators “…may be favoured with discounted tax payments deals, in
the form of prefinancing arrangements, but tax payment for local operators is mandatory”, i.e.,
tax rates could be brokered through negotiation between commercial entities and the
authority demanding the tax.
331. To get around the difficulty of using inconsistent information from the examples documented
to identify a global value for fees and licences, a proxy amount expressed as a rate of tax was
estimated for the probable value of fees and licences. Because information was often from
areas outside the UAI, assumptions had to be made about whether the cost of a fee or licence
outside the UAI was likely to be similar to those within the UAI.
332. Although there was probably no coltan, tin and tungsten production in Ituri, it is likely there
was some trade and export to Uganda through Ituri. A proxy tax for fees and licences was
therefore estimated for these resources for Ituri (along with a proportion of quantity of these
resources).
333. The proxy tax is assumed to include fees and licences paid by anyone involved in the value
chain, such as miners, porters, small traders, large traders, and exporters. See Table 7 (these
data are the same as those in Table 5).
Table 4.7: Est. proxy tax rate for value of fees and licences
Resource Ituri (%) Non-Ituri (%)
Gold 5.0 2.0
Diamonds 5.0 2.0
Coltan 5.0 2.0
Tin 5.0 2.0
Tungsten 5.0 2.0
Timber 1.0 1.0
Coffee 1.0 0.5
5.3 Tax on sales, exports and other value
334. There is evidence that tax rates on sales, exports and other value were (a) in place from August
1998, and (b) tax collection occurred systematically, if not evenly, throughout the UAI.
335. From August 1998 until the RCD started to fracture in early 1999, it maintained tax collection
structures previously in place and “…never abolished current Congolese export and import duty
rates and kept the ‘système déclaratif’ whereby traders were supposed to declare the exact
98
nature of their merchandise and pay a percentage of its value in tax; but in fact, controls were
often lax or non-existent for certain traders” (Johnson and Tegera 2007: 18).
336. The Porter Commission (2002) concluded after talking to Congolese leaders of armed forces
operating within the UAI and after hearing from Ugandan witnesses, that taxes on the sale or
export of resources were systematically applied in both Ituri and non-Ituri:
336.1 “…taxation was at the root of funding for the [Congolese] movements, and one would
expect every effort to be made to collect as much as possible, whether for personal
gain, or to finance the movements” (p.77).
336.2 “…there is no doubt that both RCD and UPDF soldiers were imposing a gold tax”
(p.197).
336.3 “There is no doubt that as a matter of practice “Security/Intelligence Funding” was
imposed on RCD, businessmen and companies, or that General Kazini’s regret was that
his commanders were likely to take the money for themselves, rather than accounting
to him” (p.199).
336.4 Victoria Company, which operated in the DRC, “deals in diamonds, gold and coffee
which it purchases from Isiro, Bunia, Bumba, Bondo, Buta and Kisangani” and “pays
taxes to MLC to back up what the Army Commander, Major General Kazini, terms ‘the
effort in the armed struggle’”. In fact, Bunia is in Ituri, so any taxes paid there were
most likely paid to RCD-ML or temporarily to the FLC (p.82).
336.5 In examining La Conmet company’s export of coltan from Beni-Lubero territories in
northern North Kivu in early 2000, the Porter Commission stated it obtained “receipts
for taxes paid by the company to the Congolese authorities in respect of that export”.
RCD-Kisangani consituted the “Congolese authorities” at the time, and this example
demonstrates that it was extracting taxes. While this example is from outside Ituri, it
is probable similar practices were followed by RCD factions within Ituri (p.182).
336.6 The Porter Commission (2002: 55) also notes “Such documentation as this Commission
has seen indicates that timber cut in the Democratic Republic of Congo is dutiable
there on export, and that such duties are levied by the rebel authorities and paid”.
337. Taking into account limitations on information and probable low and high tax rates as outlined
in Annex 4, a rate of tax was estimated to enable calculation of value extracted from resources.
Table 8 provides the rates of tax adopted for this report, as well as the range of tax noted in
Annex 4 (the data are the same as those in Table 5):
Table 4.8: Tax range and adopted tax on value
Resource
Tax Range Reported (%)
(See Annex 4)
Adopted Taxes on Value
Ituri (%) Non-Ituri (%)
Gold 28-40 28.0 5.0
Diamonds 4-15 20.0 5.0
Coltan 5-40 20.0 5.0
Tin 5-50 20.0 5.0
Tungsten n/a 20.0 5.0
Timber 6 8.0 1.0
Coffee 7 8.0 1.0
99
Gold 338. The tax on gold value in Ituri was made at the low end of the range because the
range’s minimum is far higher than for other resources, and it is not clear why
gold would be so different to other resources. The high end of the range, while
similar to the high end of the range for coltan and tin, is for 2010, seven years
after the relevant period. 28% is a conservative confident estimate.
339. Outside Ituri there is evidence that many Congolese forces targeted efforts to
exploit value on gold, so the rate of tax put in place by any group is likely to have
been more than other resources. However, gold within areas where UPDF
personnel were located was probably concentrated in the Kilo-Moto deposit
extending into Haut-Uélé (e.g., around Durba) and around Bafsawende, with a
few less significant (in terms of quantity and value) exceptions. For this reason,
the funds extracted through a tax on value imposed by UPDF personnel is
estimated to be low.
Diamonds 340. The tax on diamonds is estimated at 20% even though the reported taxes range
from 4-15%, because the rate is unlikely to be less than minerals, which are
higher than this range. There is no apparent reason for a tax on diamonds to be
less than for gold, but there is also no evidence that it was the same as for gold.
Thus, while the tax rate for diamonds may have been more, I cannot be sure.
20% is a conservative confident estimate.
Coltan,
Tin,
Tungsten
341. The tax on value for these three minerals was set at the same rate. This is
reasonable given coltan and tin, were often found in the same deposit, and
there is no reason to think that tungsten would have been taxed differently.
The rate of 20% is a conservative confident estimate within the reported range.
There is insufficient information from non-Ituri to estimate tax on value imposed
by UPDF personnel at a rate different to gold and diamonds, so the rate has
been fixed at the same level.
Timber 342. The reported tax on timber of 6% was only for exports from North Kivu in 2006
(Johnson and Tegera 2007) and does not include any other taxes on value. The
working estimate was increased to 8% to include the probability that during the
context of conflict from 1998-2003, other taxes on value were also levied, such
as at the point of production, trade or while in transit.
Coffee 343. The reported tax on coffee of 7% was only for exports from North Kivu in 2006
(Johnson and Tegera 2007) and does not include any other taxes on value. The
working estimate was increased to 8% to include the probability that during the
context of conflict from 1998-2003, other taxes on value were also levied, such
as at the point of production, trade or while in transit.
344. Table 9 shows the value in 2020 USD extracted by each method of exploitation for each
resource, across Ituri and non-Ituri, I.e., it uses the data in Table 4 disaggregated by extraction
method tax rate in Table 5. The calculations behind these amounts are in the tables in Annex 5.
100
Table 4.9: Value of exploitation disaggregated by method, Ituri and non-Ituri, 2020 USD
Theft Fees & Licences Tax of Value Total
Ituri Non-Ituri Ituri Non-Ituri Ituri Non-Ituri Ituri Non-Ituri
Gold 4,540,894 2,219,993 4,313,849 2,175,593 24,157,555 5,438,982 33,012,298 9,834,567
Diamonds 176,330 335,027 167,513 1,340,107 670,054 3,350,268 1,013,897 5,025,403
Coltan 10,963 20,830 10,415 83,320 41,660 208,299 63,038 312,449
Tin 7,523 14,294 7,147 57,176 28,588 142,939 43,258 214,409
Tungsten 2,398 4,557 2,279 18,228 9,114 45,571 13,791 68,356
Timber 516,322 129,080 252,998 258,161 2,023,982 258,161 2,793,301 645,402
Coffee 206,515 0 204,450 240,935 1,635,602 481,869 2,046,568 722,804
Total 5,460,945 2,723,781 4,958,651 4,173,520 28,566,555 9,926,089 38,986,151 16,823,390
101
Annex 1: Terms of Reference
The ICJ provided the following terms of reference (TOR) to guide this report:
(1) An expert opinion shall be obtained, which will be entrusted to four independent experts
appointed by Order of the Court after hearing the Parties.
(2) For the purposes of determining the reparation owed to the Democratic Republic of the Congo
by Uganda for the injury caused as a result of the breach by Uganda of its international obligations,
as determined by the Court in its 2005 Judgment, the Court continues to examine the full range of
claims and defences to the heads of damage claimed by the Applicant. However, with respect to
some of these heads of damage, namely, loss of human life, loss of natural resources and property
damage, the Court considers it necessary to arrange for an expert opinion, in accordance with
Article 67, paragraph 1, of its Rules. The terms of reference for the experts referred to in point (1)
above will be as follows:
II. Loss of natural resources
(a) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the
approximate quantity of natural resources, such as gold, diamond, coltan and timber,
unlawfully exploited during the occupation by Ugandan armed forces of the district of Ituri in
the relevant period?
(b) Based on the answer to the question above, what is the valuation of the damage suffered
by the Democratic Republic of the Congo for the unlawful exploitation of natural resources,
such as gold, diamond, coltan and timber, during the occupation by Ugandan armed forces of
the district of Ituri?
(c) Based on the evidence available in the case file and documents publicly available,
particularly the United Nations Reports mentioned in the 2005 Judgment, what is the
approximate quantity of natural resources, such as gold, diamond, coltan and timber,
plundered and exploited by Ugandan armed forces in the Democratic Republic of the Congo,
except for the district of Ituri, and what is the valuation of those resources?
(3) The references to the administrative divisions on the territory of the Democratic Republic of the
Congo mentioned above should be understood as those that existed in the Democratic Republic of
the Congo during the relevant period, i.e. between 6 August 1998 and 2 June 2003.
102
Annex 2: List of References
ACET (African Centre for Economic Transformation), 2014. The Coffee Value-Addition Opportunity in
Africa. http://www.oldacet.deanpaulcollins.com/wp-content/uploads/2014/08/Coffe…
All Party Parliamentary Group on the Great Lakes Region and Genocide Prevention, 2002. Cursed by
Riches: Who Benefits from Resource Exploitation in the Democratic Republic of the Congo? London:
UK Parliament.
Baker, M., R. Clausen, M. N’Goma, T. Roule, and J. Thomson, 2003. “Democratic Republic of Congo”,
Volume 3 (pp.7–115) in Conflict timber, dimensions of the problem in Asia and Africa. Report
submitted to USAID. Burlington, Vermont: ARD.
Bleischwitz, R., M. Dittrich and C. Pierdicca, 2012. Coltan from Central Africa: International Trade and
Implications for Any Certification. European Economic Policy Briefings, BEEP no.23. Bruges, Belgium:
College of Europe.
Cerbu, G. (ed.). 2016. Central Africa Congo Basin Timber. Washington D.C.: PROFOR.
https://www.profor.info/sites/profor.info/files/PROFOR_CongoTimber_web_…
Chatham House 2020. Forest Governance and Legality: Democratic Republic of Congo. Map showing
deforestation, 2001 v. 2018. https://forestgovernance.chathamhouse.org/countries/democraticrepublic-
of-the-congo
Congo European Network (Brussels), REC-info No.5/2001, 23 May 2001, Heading “North Kivu”, point
No. 15, reporting news gathered from local sources on 14 April 2001 (from the DRC Memorial
Annexe E Vol. 2)
Counsell, Simon, 2006. Forest Governance in the Democratic Republic of Congo. Brussels: FERN.
Danssaert, P. and B. J. Thomas, 2006. Greed & Guns: Uganda’s Role in the Rape of the Congo.
Antwerp: IPIS. https://ipisresearch.be/wp-content/uploads/2015/05/Greed-and-guns.pdf
Debroux, L., T. Hart, D. Kaimowitz, A. Karsenty, and G. Topa, (eds.) 2007. Forests in Post-Conflict
Democratic Republic of Congo: Analysis of a Priority Agenda. Jakarta: Center for International
Forestry Research (CIFOR).
Democratic Republic of the Congo, 2018a. Case Concerning Armed Activities on the Territory of the
Congo (Democratic Republic of the Congo v. Uganda). Second Phase. Question of Reparation. Replies
and Further Evidence Provided by the Democratic Republic of the Congo (Final Version). Volume. 26
October 2018.
Democratic Republic of the Congo, 2018b. Case Concerning Armed Activities on the Territory of the
Congo (Democratic Republic of the Congo v. Uganda). Second Phase. Question of Reparation.
Counter-Memorial of the Democratic Republic of the Congo. February 2018.
Democratic Republic of the Congo, 2016. Case Concerning Armed Activities on the Territory of the
Congo (Democratic Republic of the Congo v. Uganda). Second Phase. Question of Reparation.
Memorial of the Democratic Republic of the Congo. September 2016.
DIAR (Diamond Industry Annual Review), 2005. DR Congo 2005. https://impacttransform.org/wpcontent/
uploads/2017/09/DRC-2005.pdf
103
Dietrich, C. 2002. Monnaie Forte : L’économie criminalisée des diamants dans la République
démocratique du Congo et les pays voisins. Ottawa: Partenariat Afrique Canada.
Djiré, A. 2003. Le secteur informel du bois d’oeuvre. Rapport d’appui à la revue
du secteur forestier en RDC–Rapport technique. Montpellier : Centre
de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD).
D’Souza, Kevin, 2007. Artisanal Mining in the DRC. Briefing Note. CASM, UK.
UNECE (United Nations Economic Commission for Europe), 2020. Forest Products Conversion Factors
for the UNECE Region. ECE/TIM/DP/49.
http://www.unece.org/fileadmin/DAM/timber/publications/DP-49.pdf
European Union, 2008. The Forests of the Congo Basin. State of the Forest 2008. (Eds) C. de
Wasseige, D. Devers, P. de Marcken, R. Eba’a Atyi, R. Nasi, and P. Mayaux. Luxembourg: Publications
Office of the European Union.
FAO, 2020. Forest Products Yearbook. http://www.fao.org/3/a-y2610m.pdf
GAO (General Accounting Office), 2002. Report to Congressional Requesters: International Trade.
Critical Issues Remain in Deterring Conflict Diamond Trade. Washington DC: United States
Government (GAO-02-678). https://www.gao.gov/assets/240/234898.pdf
Garrett, N., 2008. Artisanal Cassiterite Mining and Trade in North Kivu. Implications for Poverty
Reduction and Security. Communities and Artisanal & Small-scale Mining Initiative (CASM).
Washington DC: World Bank.
Global Witness, 2005. Undermining Peace. Tin: The Explosive Trade in Cassiterite in Eastern DRC.
London, UK. https://cdn.globalwitness.org/archive/files/pdfs/under-mining%20peace.p…
Global Witness. 2002. The Logs of War: The Timber Trade and Armed Conflict. Programme for
International Co-operation and Conflict Resolution, Fafo Report 379. London.
Goreux, L., 2001. Conflict Diamonds. Africa Region Working Paper Series No.13, Washington D.C.:
World Bank.
http://documents1.worldbank.org/curated/en/216551468741661999/pdf/multi…
Hayes, K. and R. Burge, 2003. Coltan Mining in the Democratic Republic of Congo: How tantalumusing
industries can commit to the reconstruction of the DRC. Cambridge, UK: Fauna and Flora
International.
Hinton, J., 2012. “Uganda Case Study” in Analysis of Formalization Approaches in the Artisanal and
Small-Scale Gold Mining Sector Based on Experiences in Ecuador, Mongolia, Peru, Tanzania and
Uganda. Geneva: UNEP. https://www.responsiblemines.org/wpcontent/
uploads/2018/05/Case_Study_Uganda_June_2012.pdf
HRW (Human Rights Watch), 2005. The Curse of Gold: Democratic Republic of Congo. New York.
https://www.hrw.org/sites/default/files/reports/drc0505_0.pdf
HRW (Human Rights Watch), 2003. Ituri: “Covered in Blood”. Ethnically Targeted Violence in
Northeastern DR Congo. New York. https://www.hrw.org/reports/2003/ituri0703/DRC0703.pdf
HRW (Human Rights Watch), 2002. Kisangani in the Congo War. New York.
https://www.hrw.org/reports/2002/drc2/DRC0802-02.htm
104
HRW (Human Rights Watch), 2001. Uganda in Eastern DRC. Fueling Political and Ethnic Strife. New
York. https://www.hrw.org/report/2001/03/01/uganda-eastern-drc/fueling-politc…
ICG, 2004. Maintaining Momentum in the Congo: The Ituri Problem. Africa Report no.84. Brussels.
ICG, 2003. Congo Crisis: Military Intervention in Ituri. Africa Report no.64. Brussels.
ICG, 2000a. Uganda And Rwanda: Friends Or Enemies? Africa Report no.14. Brussels.
ICG, 2000b. Scramble for the Congo. Anatomy of an Ugly War. Africa Report no.26. Brussels.
International Alert, 2010. The Role of the Exploitation of Natural Resources in Fuelling and
Prolonging Crises in the Eastern DRC. London. https://www.internationalalert.
org/sites/default/files/publications/Natural_Resources_Jan_10.pdf
ICO (International Coffee Organization), 2020. Historical Data on the Global Coffee Trade.
http://www.ico.org/new_historical.asp
IPIS (International Peace Information Service), 2002. Supporting the War Economy in the DRC:
European Companies and the Coltan Trade: Five Case Studies. (Eds) M-O. Herman, D. Broederlijk and
P. Vermaerke. Antwerp.
IPIS, 2020. Interactive Minerals Map. https://ipisresearch.be/mapping/webmapping/drcongo/v5/#-
1.4045201543864891/28.801971434596567/5.774660859437679/2/1/1.9.20,2.157bpc
ITTO (International Tropical Timber Organization), 2020. DRC timber production and exports data:
https://www.itto.int/biennal_review/?mode=searchdata
Johnson, D. and A. Tegera, 2007. Rules for Sale: formal and informal cross-border trade in eastern DRC.
Goma: Pole Institute.
Johnson, Dominic and A. Tegera, 2005. Digging Deeper: How the DR Congo’s mining policy is failing the
country. Goma: Pole Institute. http://archive.niza.nl/docs/200603061556507567.pdf
Johnson, S. E., M. Adams and M. Miyake, 2004. Trends in the Tropical Timber Trade, 2002-2003. United
Nations: Geneva, ECE/TIM/DP/35.
Kamungele, K. K., 2013. “The DRC Coffee Country Profile”. Presentation for 10th African Fine Coffee
Conference & Exhibition, Kampala, Uganda. 14 February 2013.
Le Billon, P. and C. Hocquard, 2007. Filières industrielles et conflits armés: le cas du tantale dans la
région des Grands Lacs. Écologie & Politique 34: 83-92.
Lescuyer, G., E. Essiane Mendoula, R. Tsanga, 2016. Description of the artisanal supplies of wooden
products in Yaoundé, Douala and Kinshasa for the project “Congo Basin Timber.” CIFOR, Version of
27 January 2016. https://agritrop.cirad.fr/580181/1/CIFOR%20Report%20-
%20Demands%20for%20legal%20timber%20v3%20290316.pdf
Megevand, C., with A. Mosnier, J. Hourticq, K. Sanders, N. Doetinchem, and C. Streck, 2013.
Deforestation Trends in the Congo Basin: Reconciling Economic Growth and Forest Protection.
Washington D.C.: World Bank.
105
MONUC (UN Organization Mission in the Democratic Republic of the Congo), 2004. Special report on
the events in Ituri January 2002-December 2003. July, New York (S/2004/573).
Moyroud, C. and J. Katunga, 2002. “Coltan exploitation in the Eastern Democratic Republic of
Congo” in Scarcity and Surfeit: The ecology of Africa’s conflicts, (eds) J. Lind and K. Sturman. Pretoria:
African Centre for Technology Studies and Institute for Security Studies.
Mthembu-Salter, G., 2015b. “Baseline study three: Production, trade and export of gold in Orientale
Province, Democratic Republic of Congo”. 9th Multi-stakeholder forum on responsible mineral supply
chains, 4-6 May 2015. Paris, France. https://www.oecd.org/daf/inv/mne/Gold-Baseline-Study-3.pdf
Mthembu-Salter, G., 2015a. “Baseline study four: Gold trading and export in Kampala, Uganda. 9th
Multi-stakeholder forum on responsible mineral supply chains”, 4-6 May 2015. Paris, France.
https://www.oecd.org/daf/inv/mne/Gold-Baseline-Study-4.pdf
Nest, Michael 2011. Coltan. London: Polity Press.
OHCHR (United Nations Office of the High Commissioner for Human Rights), 2003. Report of the
Mapping Exercise. Geneva. https://www.securitycouncilreport.org/atf/cf/%7B65BFCF9B-6D27-
4E9C-8CD3-CF6E4FF96FF9%7D/ROL%20OHCHR%20Report%20Aug%202010%20DRC.pdf
PAC (Partenariat Afrique Canada), 2004. Revue annuelle de l’industries des diamants : République
Démocratique du Congo 2004. https://impacttransform.org/wp-content/uploads/2017/09/RDC-
2004.pdf
Porter Commission, 2002. Judicial Commission of Inquiry into Allegations into Illegal Exploitation of
Natural Resources and Other Forms of Wealth in the Democratic Republic of Congo 2001. Kampala:
Republic of Uganda, November 2002.
Price, S., Donovan, D., & De Jong, W. I. L., 2007. “Confronting conflict timber” in Extreme Conflict and
Tropical Forests (pp. 117-132). Netherlands: Springer.
Raeymaekers, T., 2002. Network War: An Introduction to Congo's Privatised War Economy. Antwerp:
International Peace Information Service.
Redmond, I. 2001. Coltan Boom. Gorilla Bust. The impact of coltan mining on gorillas and other
wildlife in Eastern DR Congo. Atlanta: Dian Fossey Gorilla Fund and Horsham, UK: Born Free
Foundation.
Salter, G., 2008. The Impact of Mining in Kalima. Pretoria: Institute of Security Studies.
Spittaels, S. and F. Hilgert, 2013. Analysis of the interactive map of artisanal mining areas in Eastern
DR Congo. Antwerp: International Peace Information Service.
Spittaels, S. and F. Hilgert, 2010. Mapping Conflict Motives: Province Orientale (DRC). Antwerp:
International Peace Information Service. https://ipisresearch.be/wpcontent/
uploads/2010/03/20100513_provinceorientale-2.pdf
Tegera, A. and S. Mikolo, 2002. The Coltan Phenomenon. How a rare mineral has changed the life of
the population of war-torn North Kivu province in the East of the Democratic Republic of Congo.
Goma: Pole Institute.
106
Uganda National Environmental Management Authority, 2019. Artisanal and Small Scale Gold
Mining. Television documentary. https://web.unep.org/globalmercurypartnership/artisanal-andsmall-
scale-gold-mining-uganda
Uganda, 2016. Case Concerning Armed Activities on the Territory of the Congo. Democratic Republic
of the Congo v. Uganda. Memorial of Uganda on Reparations. Volume 1. 28 September 2016.
Uganda, 2018. Case Concerning Armed Activities on the Territory of the Congo. Democratic Republic
of the Congo v. Uganda. Counter-Memorial of Uganda on Reparations. Volume 1. 6 February 2018.
Umunay, Peter, 2011. Illegal Logging in Congo Basin - Ituri Case Study. Global Forest Atlas.
Umunay, P. and J. R. Makana, 2009. Etude sur le Commerce Transfrontalier de Bois Dans le Nord-est
de la République Démocratique du Congo. Unpublished report submitted to IUCN DRC.
UN ComTrade, 2020. Export and Import data website: https://comtrade.un.org/data
UNPE (UN Panel of Experts), 2003. Final Report of the UN Panel of Experts on the Illegal Exploitation
of Natural Resources in the DR Congo. October, New York. (S/2003/1027)
UNPE (UN Panel of Experts), 2002a. Report of the UN Panel of Experts on the Illegal Exploitation of
Natural Resources in the DR Congo. October, New York. (S/2002/1146)
UNPE (UN Panel of Experts), 2002b. Interim Report of the UN Panel of Experts on the Illegal
Exploitation of Natural Resources in the DR Congo. October, New York. (S/2002/565)
UNPE (UN Panel of Experts), 2001a. Report of the UN Panel of Experts on the Illegal Exploitation of
Natural Resources in the DR Congo. April, New York. (A/2001/357)
UNPE (UN Panel of Experts), 2001b. Addendum to the Report of the UN Panel of Experts on the
Illegal Exploitation of Natural Resources in the DR Congo. November, New York. (S/2001/1072)
USGS, various years. Minerals Yearbooks for gold, diamonds, tantalite-niobium, tin, and tungsten.
https://www.usgs.gov/centers/nmic/commodity-statistics-and-information
US Inflation Calculator, 2020. Based on US Government CPI data published on October 13, 2020
based on US Labor Dept Bureau of Labor Statistics data: https://www.usinflationcalculator.com
Wilkins, K., 2019. “The Cocoa and Coffee Opportunity in the Democratic Republic of the Congo. A
guide to the Congolese cocoa and coffee market for businesses, buyers and investors”. ELAN RDC.
107
Annex 3: Commodity codes
Searches on the UN ComTrade database used the following codes listed below.
Gold 710812 - Metals; gold, non-monetary, unwrought (but not powder).
Coltan 261590 - Niobium, Tantalum, vanadium ores and concentrates.
Tin 260900 - Tin ores and concentrates.
Tungsten 261100 - Tungsten ores and concentrates.
Timber 4407 - Wood sawn or chipped lengthwise, sliced, peeled, whether or not planed,
sanded or finger jointed, of a thickness exceeding 6mm.
Coffee 090111 - Coffee; not roasted or decaffeinated.
In regard to diamonds, ComTrade data were not used to identify quantities or value likely to have
been produced in UAI for two reasons:
• Data for weight both industrial and gem diamonds are absent in most cases, so quantity
cannot be established for either kind of diamond; and
• Even though value is recorded, because weight is absent price per carat cannot be calculated
and therefore could not be used to inform this report.
ComTrade codes for industrial and non-industrial diamonds are therefore not included because they
were not used.
108
Annex 4: Reported taxes on natural resources
This table summarised reports of taxes on value, profits and exports in case file and other
documents.
Resource/
Source
Rate
Reported
Tax on what? Collector Period Region
Gold
(UNPE 2001a: §59)
1gram/day
(approx.
28%)
Daily ‘fee’ ‘Ugandan local
commanders and
some of the
soldiers’
Late
1999?
Haut-Uélé/Ituri
border area.
Gold (International
Alert 2010: 43)
40% Export value Civilian
Administration
2010 Ituri
Gold (Johnson and
Tegera 2007: 94)
40%
(30%
OKIMO)
(10% Prov.)
Tax on ore
leaving mine
OKIMO royalties;
Ituri Admin.
2006? Ituri
Gold (Johnson and
Tegera 2007: 94)
$1,392 Fees and
licences to
produce
All authorities 2006? Ituri
Gold (Johnson and
Tegera 2007: 87)
4.75% Export taxes All authorities 2006? South Kivu
Gold (Johnson and
Tegera 2007: 90)
$75,000 Export
licence fee
Paid in Kinshasa 2006? All DRC
Gold, industrial
(Johnson and Tegera
2007: 24)
3% Exports OFIDA 2007? North Kivu
Gold, artisanal
(Johnson and Tegera
2007: 24)
1.5% Exports OFIDA 2007? North Kivu
Diamonds (RPE
2001a: §127)
5% Export value ‘Congo Desk’
Rwanda
1998-
2001
Kisangani
comptoirs
Diamonds (Johnson
and Tegera 2005: 97)
4% Value DRC Government 2004 Nationwide
Diamonds (UNPE
2001b: §46.)
15% Export value Rwanda; RCDGoma
2001 From the DRC
(5%) Export value ‘Congo Desk’, Rw 2001 From the DRC
(10%) Export value ‘Rebel Admin.’ 2001 From the DRC
Diamonds, industrial
(Johnson and Tegera
2007: 24)
3% Exports OFIDA 2007? North Kivu
Diamonds, artisanal
(Johnson and Tegera
2007: 24)
1.5% Exports OFIDA 2007? North Kivu
Coltan
(IPIS, 2002: 10)
8% Exports by
comptoirs
RCD; RCD-Goma 1998-
2000
South Kivu
40%?
($10/kg)
Exports by
SOMIGL
RCD-Goma From
Nov. 2000
South Kivu; North
Kivu
Coltan
(Le Billon and
Hocquard, 2007)
7% Profits RCD-Goma Circa
2000
South Kivu; North
Kivu
11% Profits ‘Armed Groups’ Circa
2000
South Kivu; North
Kivu
22% Of profits
spent on
RCD-Goma Circa
2000
South Kivu; North
Kivu
109
‘licences &
Fees’
Coltan (Johnson and
Tegera 2005: 37)
11% total Value
($10/kg
reported)
All authorities 2005 Mumba/Bibatama
mine, North Kivu
(7%) Commune
(2%) Zone
(2%) ‘Mining Division’,
Goma
Coltan (Johnson and
Tegera 2002: 7)
$20,000 per
month
Coltan
exports from
the DRC
RCD; RCD-Goma 1998-Nov
2000
South Kivu; North
Kivu?
Coltan (Johnson and
Tegera 2002: 7)
$1,124,970 Coltan (and
cassiterite?)
SOMIGL to RCD Dec 2000
alone
Rwandan area of
influence
Coltan (Redmond
2001: 11)
$4/kg
(approx.
17% tax)
Tax on
exports by
kilogram
Comptoirs to ?
(RCD? RPA?)
April-May
2001
South Kivu; North
Kivu
Coltan (Redmond
2001: 11)
$7.50/kg
(approx.
32% tax)
Weekly fee to
work in
mines
Paid by miners to
(1) military and
(2) ‘Chef de
colline)
April-May
2001
Kahuzi-Biéga NP,
South Kivu
Coltan
(Congo European
Network 2001)109
$6/kg
(approx.
25% tax)
Export tax on
value (plus
$40,000
annual fee to
export).
RCD-Goma Nov 2000-
April 2001
From the DRC
$4/kg for
more than
15mt
(approx.
17% tax)
Export tax on
consignments
more than
15mt
RCD-Goma Nov 2000-
April 2001
From the DRC
Coltan (Johnson and
Tegera 2005: 47)
Fixed
royalty tax
of $5,000
On coltan
miners
RCD-Goma Pre-2004 South Kivu; North
Kivu
Coltan/cassiterite?
(Johnson and Tegera
2005: 57)
$1-$1.50/kg
(approx.
5%)
Value ‘Military Forces’ 1998-
2005
Walikale Territory
(Goma/Bukavu-
Kisangani road)
‘Minerals’
(coltan/ cassiterite?)
(International Alert
2010: 43)
15% ‘Value’
($365/mt)
Civilian Admin.
(b/w Bisie-Goma)
2010 Western North
Kivu within
Rwandan area of
influcence
Cassiterite
(Garrett 2008: 32)
10% $4/kg + 10%
of minerals
carried
FARDC 2008 Bisie mine,
western North
Kivu
Cassiterite (Johnson
and Tegera 2005: 47)
$2,500 fee Imposed on
traders
RCD Pre-2004 South Kivu; North
Kivu
Cassiterite (Johnson
and Tegera 2005: 59)
50% Quantity RCD 2004 Bisie mine, North
Kivu
Cassiterite, artisanal
(Johnson and Tegera
2007: 24)
15% Value North Kivu
authorities
2003 North Kivu
109 Reported in DRC (2016) Memorial Annex E, Vol. 2, p.15.
110
Cassiterite, artisanal
(Johnson and Tegera
2007: 24)
10% Prod. kept by
mine
authorities.
North Kivu
authorities
2003 North Kivu
Cassiterite,
industrial (Johnson
and Tegera 2007: 24)
10% Taxed before
leaving mine
North Kivu
authorities in
Mine (OFIDA)
2003 North Kivu
Timber (Johnson and
Tegera 2007: 24)
6% Export tax on
untreated
timber
North Kivu
authorities
2003 North Kivu
Green coffee beans
(Johnson and Tegera
2007: 60)
7% “Total”
export taxes
North Kivu
authorities
2003 North Kivu
111
Annex 4.5: Calculations of Value
A4.5.1 Gold
Table A4.5.1.1: Quantity, kilograms 1998* 1999 2000 (a) 2001 2002 2003* Total
*1998 and 2003 five months only (Jan-Jun) (Jul-Dec)
D. R. Congo - Production
1 Formal production (b) 62.90 207.00 26.00 26.00 6,100.00 7,600.00 1,708.33 15,730.23
2 Assume 80% of L1 from non-
Government area (c)
50.33 165.60 20.80 20.80 4,880.00 6,080.00 1,366.67 12,584.20
3 75% of L2 in UAI to June 2000;
70% in UAI from July 2000 (d)
37.75 124.20 15.60
14.56 3,416.00 4,256.00 956.67 8,820.78
4 Add est. national artisanal
production (e)
2,083.33 5,000.00 2,500.00 2,500.00 5,000.00 5,000.00 2,083.33 24,166.66
5 80% of L4 in non-Govt held 1,666.67 4,000.00 2,000.00 2,000.00 4,000.00 4,000.00 1,666.67 19,333.34
6 75% of L6 in UAI to June 2000;
70% in UAI from July 2000 (d)
1,250.00 3,000.00 1,500.00 1,400.00 2,800.00 2,800.00 1,166.67 13,916.67
7 Total Est. UAI Production
(R3 + R6)
1,287.75 3,124.20 1,515.60 1,414.56 6,216.00 7,056.00 2,123.33 22,737.44
a. 2000 is split into two six-month periods to reflect Uganda’s loss of influence in Kisangani after June 2000. Loss of influence reduced the ability of
UPDF personnel to extract value from gold in Kisangani.
b. Based on USGS data (most recent Yearbook)
c. See text for explanation.
d. Est. UAI share was 75% of non-government held area to June 2000, then 70% from July 2000.
e. See text for explanation. Base estimate used was 5,000 kg per year for the DRC revised accordingly for non-government-held area, then UAI.
112
Table A4.5.1.2: DRC Congo gold
exports
1998* 1999 2000 (a) 2001 2002 2003* Total
*1998 and 2003 five months only (Jan-Jun) (Jul-Dec)
D. R. Congo - Exports
1 Formal exports (b) 419.58 241.56 412.50 412.50 887.00 527.00 1.25 2,901.39
2 Assume 80% of L1 from non-Govt
area (c)
335.67 193.25 330.00 330.00 709.60 421.60 1.00 2,321.12
3 Est. formal exports from UAI:
75% of L2 to June 2000; 70%
from July 2000 (d)
251.75 144.94 247.50 231.00 496.72 295.12 0.70 1,667.73
4 Est. UAI production from L7,
Table A4.5.1.1
1,287.75 3,124.20 1,515.60
1,414.56 6,216.00 7,056.00 2,123.33 22,737.44
5 UAI production minus exports
(L4 - L3), i.e., smuggled gold.
1,036.00 2,979.26 1,268.10 1,183.56 5,719.28 6,760.88 2,122.63 22,069.71
a. 2000 is split into two six month periods to reflect Uganda’s loss of influence in Kisangani after June 2000.
b. Based on ComTrade import data for “All” reporters.
c. See text for explanation.
d. Est. UAI share was 75% of total non-government held area to June 2000, then 70% from July 2000.
113
Table A4.5.1.3: Uganda gold
production and exports
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
Uganda - Production and Exports
1 Formal production (a) 3.33 5.00 56.00 0.00 3.00 16.67 84.00
2 Est. artisanal production (b) 416.67 1,000.00 1,000.00 1,000.00 1,000.00 416.67 4,833.34
3 Est. total production (L1 + L2) 420.00 1,005.00 1,056.00 1,000.00 1,003.00 1,040.00 5,524.00
4 Formal exports (c) 936.25 4,231.00 5,297.00 6,161.00 7,117.00 1,449.17 25,191.42
5 Exports surplus to production:
assume from UAI (L4 - L3) (d)
516.25 3,226.00 3,241.00 5,161.00 6,114.00 1,015.83 19,274.08
a. Based on USGS data (most recent Yearbook)
b. See text for explanation of 1,000 kg estimate per year.
c. Based on Uganda Bureau of Statistics data in Table 8.2 in Case Concerning Armed Activities on the Territory of the Congo. Democratic Republic of the
Congo v. Uganda. Counter-Memorial of Uganda on Reparations. Volume 1. 6 February 2018 (Uganda).
d. Uganda exports excess to production assumed to be UAI-origin because during the 1998-2003 period: cross-border trade in gold between Uganda
and either Rwanda or Burundi unlikely; Kenya produced and exported gold, but no reason for traders to bring DRC gold to export from Kenya if
possible from Uganda; Central African Republic production unlikely to have transited through the DRC to Uganda; and Sudanese production unlikely
to have been exported via Uganda.
114
Table A4.5.1.4: UAI smuggled gold v.
Ugandan export ‘surplus’.
*1998 and 2003 5mth only
1998* 1999 2000 2001 2002 2003* Total
Comparison of UAI smuggled gold v. Uganda exports surplus to production (a)
1 UAI smuggled gold (Table A4.5.1.2: L5) 1,036.00 2,979.26 2,451.66 5,719.28 6,760.88 2,122.63 22,069.71
2
Ugandan exports surplus to production
(Table A4.5.1.3: L5)
516.25 3,226.00 3,241.00 5,161.00 6,114.00 1,015.83 19,274.08
3
Take highest yearly est. from L1 or L2
(b)
1,036.00
(from DRC)
3,226.00
(from Ug.)
3,241.00
(from Ug.)
5,719.28
(from DRC)
6,760.88
(from DRC)
2,122.63
(from DRC)
22,105.79
4 Est. quantity in UAI: 1,036.00 3,226.00 3,241.00 5,719.28 6,760.88 2,122.63 22,105.79
a. The difference between DRC smuggled gold and Ugandan exports surplus to production is assumed to be that portion of gold from UAI that
transitted through Uganda to the international market but was not captured in any statistics.
b. Because the difference between the DRC and Ugandan data cannot be reconciled, and given that both sets of data are based on conservative
estimates of informal production and trade, it was reasonable to taken the highest yearly estimate from either L1 or L2 as the likely quantity
smuggled from UAI into Uganda.
115
Gold: Value, USD
Table A4.5.1.5
*1998 and 2003 five months only
1998* 1999 2000 (a) 2001 2002 2003* Total
1 Est. quantity from UAI (b) 1,036.00 3,226.00 3,241.00 5,719.28 6,760.88 2,122.63 22,105.79
2 Est. price, USD per kg (c) 6,145.88 5,821.54 5,832.62 5,664.18 6,471.68 7,592.64
3 Total (L1 x L2) (d) 6,367,132 18,780,298 18,903,518 32,395,057 43,754,248 16,116,398 136,316,651
4 To get 2020 USD multiply L3
by … (e)
1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD
(L3 x L4)
10,187,411 29,297,264 28,544,312 47,620,734 63,443,660 22,724,121 201,817,503
a. Jan-Jun and Jul-Dec periods for 2000 have been merged back into a single year.
b. From L5 in Table A4.5.1.4.
c. Prices based on World Gold Council price database annual averages, accessed on 6 December 2020: From World Gold Council:
https://www.gold.org/goldhub/data/gold-prices. Annual price was then reduced by 35% to better reflect probable price at points of opportunities
for exploitation in the DRC. This base price and the original adopted price are shown in Table 2.
d. Total figures rounded-up (no cents included).
e. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
116
Gold: Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.1.6
Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kilograms) 9,948 45% 12,158 55% 22,106
2 Base value of quantity (b) 90,817,876 45% 110,999,627 55% 201,817,504
3 Est. value of Theft (c) 4,540,894 5.0% 2,219,993 2.0% 6,760,887
4 Est. Fees & Licences (d) 4,313,849 5.0% 2,175,593 2.0% 6,489,442
5 Est. of Taxes on Value (e) 24,157,555 28.0% 5,438,982 5.0% 29,596,537
6 Total est. value of damages $ 33,012,298 $ 9,834,568 $ 42,846,866
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
117
A4.5.2 Diamonds
Table A4.5.2.1: Quantity, carats 1998* 1999 2000 (a) 2000 (a) 2001 2002 2003* Total
*1998 and 2003 five months only (Jan-Jun) (Jul-Dec)
D. R. Congo
1 Est. DRC production (b) 10,833,333 20,101,999 7,950,000 7,950,000 16,902,001 21,802,002 10,833,333 96,372,668
2 Assume 9% of L1 from
Equateur and Orientale (c)
975,000 1,809,180 715,500 715,500 1,521,180 1,962,180 975,000 8,673,540
3 70% of L2 in UAI to end June
2000 (d)
682,500 1,266,426 500,850 2,449,776
4 35% of L2 in UAI from July
2000 (d)
250,425 532,413 686,763 341,250 1,810,851
5 Est. quantity in UAI (L3 + L4) 682,500 1,266,426 500,850 250,425 532,413 686,763 341,250 4,260,627
a. 2000 is split into two six month periods to reflect Uganda’s loss of influence in Kisangani after June 2000.
b. Based on PAC (Partenariat Afrique Canada), Revue annuelle de l’industries des diamants: République Démocratique du Congo 2004, Table 1.
(Production data exclude Sengamines’ production in government-held territory). https://impacttransform.org/wp-content/uploads/2017/09/RDC-
2004.pdf.
c. Christian Dietrich in Monnaie Forte: L’économie criminalisée des diamants dans la République démocratique du Congo et les pays voisins (Ottawa:
Partenariat Afrique Canada: 2) estimates that 10% of DRC diamond production is from Equateur and Orientale. To allow for a margin of error, this
amount was ‘discounted’ by 10%. I.e., this report estimates 9% of national production occurred in these provinces.
d. See text for explanation
118
Diamonds: Value, USD
Table A4.5.2.2 1998* 1999 2000 (a) 2000 (a) 2001 2002 2003* Total
*1998 and 2003 five months only (Jan-Jun) (Jul-Dec)
D. R. Congo
1 Est. quantity in UAI (a) 682,500 1,266,426 500,850 250,425 532,413 686,763 341,250 4,260,627
2 Est. price, USD per carat (b) 12.09 8.16 9.32 9.32 12.21 12.56 17.83
3 Total (L1 x L2) 8,248,652 10,330,063 4,667,213 2,333,606 6,500,912 8,628,508 6,085,327 46,794,281
4 To get 2020 USD multiply L3
by … (c)
1.60 1.56 1.51 1.51 1.47 1.45 1.41
5
Est. total value in 2020 USD
(L3 x L4)
13,197,844 16,114,898 7,047,491 3,523,746 9,556,341 12,511,336 8,580,311 70,531,967
a. From L5 in previous table.
b. Est. based on an annual average price for diamonds produced artisanally (the method in Equateur and Orientale), calculated by dividing artisanal
value by artisanal quantity in Tableau 1, PAC (Partenariat Afrique Canada), Revue annuelle de l’industries des diamants: République Démocratique
du Congo 2004.
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
119
Diamonds: Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.2.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (carats) 213,031 5% 4,047,596 95% 4,260,627
2 Base value of quantity (b) 3,526,598 5% 67,005,369 95% 70,531,967
3 Est. value of Theft (c) 176,330 5.0% 335,027 0.5% 511,357
4 Est. Fees & Licences (d) 167,513 5.0% 1,340,107 2.0% 1,507,620
5 Est. of taxes on Value (e) 670,054 20.0% 3,350,268 5.0% 4,020,322
6 Total est. value of damages $ 1,013,897 $ 5,025,402 $ 6,039,299
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
120
A4.5.3 Coltan
Table A4.5.3.1: Quantity, kilograms
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
Uganda
1 Est. production (a) 0 0 3,000 11,000 6,000 6,667 26,667
2 Est. exports (b) 4,593 800 6,692 4,038 0 2,542 18,665
3 Exports surplus to production; assume
from DRC (L1 - L2) (c)
4,593 800 3,692 (6,962) (6,000) (4,125) (8,002)
D. R. Congo
4 Est. DRC exports (d) 9,453 1,875 231,452 44,073 73,971 74,405 435,229
5 Est. 95% of L4 from non-Govt area (e) 8,980 1,781 219,879 41,869 70,272 70,685 413,466
6 Est. 5% of L5 from UAI (f) 449 89 10,994 2,093 3,514 3,534 20,673
Other probable exporters of DRC coltan
7 Est. exports from Kenya (g) 0 0 0 22,078 60,903 17,748 100,729
8 Assume 50% of L7 via UAI (h) 0 0 0 11,039 30,452 8,874 50,365
9 Est. exports from Central African. Rep. (g) 0 0 0 0 9,909 0 9,909
10 Assume 100% of L9 via UAI (h) 0 0 0 0 9,909 0 9,909
11 Est. exports from Congo-Brazzaville (i) 73,086 0 0 0 0 0 73,086
12 Assume 33% of L11 via UAI (j) 24,362 0 0 0 0 0 24,362
13 Total est. additional (L8 + L10 +L12) 24,362 0 0 11,039 40,351 8,874 84,626
14 Est. quantity from UAI (L3 + L6 + L13) 29,404 889 14,686 6,170 37,875 8,283 97,307
a. Ugandan production from USGS (most recent Mineral Yearbook). USGS data for niobium and tantalum was combined for period of interest.
b. Based on either import or export data from ComTrade (no transaction was counted twice; different reporters for different years).
c. Uganda exports in excess of production assumed to be of DRC origin because the only other producers nearby were Rwanda and Burundi, and
cross-border trade in coltan was unlikely 1998-2003. Zimbabwe also produced coltan from 2001, but this would not be exported via Uganda.
d. Based on either import or export data from ComTrade (no transaction was counted twice; different reporters for different years).
e. See text for explanation of estimate that 95% of coltan exports came from non-Government area.
f. See text for explanation of estimate that 5% of coltan production in non-Government area was in UAI.
g. Based on ComTrade. Kenya and Central African Republic did not produce coltan 1998-2003, so assumed these originated in the DRC.
h. We cannot know if Kenya and CAR exports originated in state-held or UAI areas, so assumed 50% only may have originated in UAI.
i. Congo-Brazzaville and its neighbours are not producers. Assumed, therefore, that ComTrade imports reported as from “Congo” were from the DRC.
j. Because I cannot know by what route Congo-Brazzaville exports left the DRC, I assumed 33% only passed via Uganda.
121
Coltan (Niobium-Tantalite): Value, USD
Table A4.5.3.2
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
1 Est. quantity from UAI (a) 29,404 889 14,686 6,170 37,875 8,283 97,307
2 Est. price, USD per kg (b) 8.44 31.14 74.50 55.07 30.71 9.17
3 Total (L1 x L2) 248,080 27,681 1,094,149 339,835 1,162,963 75,971 2,948,679
4 To get 2020 USD multiply L3 by … (c) 1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD (L3 x L4) 396,929 43,182 1,652,165 499,558 1,686,296 107,120 4,385,250
a. From L14 in previous table.
b. Initial price estimate based on an average of all price observations for each year (1998-2003) available from ComTrade records for niobium-tantalite
imports and exports involving East and Central Africa. This price was then reduced by 35% to better reflect probable price at points of opportunities
for exploitation in the DRC (always less than prices paid by international importers).
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
122
Coltan (Niobium-Tantalite): Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.3.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kgs) 4,204 5% 79,878 95% 84,082
2 Base value of quantity (b) 219,263 5% 4,165,988 95% 4,385,250
3 Est. value of Theft (c) 10,963 5.0% 20,830 0.5% 31,793
4 Est. Fees & Licences (d) 10,415 5.0% 83,320 2.0% 93,735
5 Est. of taxes on Value (e) 41,660 20.0% 208,299 5.0% 249,959
6 Total est. value of damages $ 63,038 $ 312,449 $ 375,487
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
123
A4.5.4 Tin
Table A4.5.4.1
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
Uganda
1 Est. production (a) 459 333 333 18,000 0 417 19,542
2 Est. exports (b) 0 0 0 46,897 1,500 4,175 52,572
3
Exports surplus to production; assume
from DRC (L1 - L2) (c)
0 0 0 28,897 1,500 3,758 34,155
D. R. Congo
4 Est. DRC exports (d) 161,015 192,750 278,761 2,823,640 413,840 328,428 4,198,434
5 Assume 95% of L4 from non-Govt area (e) 152,965 183,113 264,823 2,682,458 393,148 312,007 3,988,514
6 Assume 5% of L5 from UAI (f) 7,648 9,156 13,241 134,123 19,657 15,600 199,425
Other probable exporters of DRC cassiterite
7 Est. exports from Tanzania (g) 0 0 0 0 10,000 5,592 15,592
8 Assume 50% of L7 via UAI (h) 0 0 0 0 5,000 2,796 7,796
9 Est. exports from Congo-Brazzaville (i) 102,708 225,000 462,743 412,029 351,174 393,500 1,947,154
10 Assume 33% of L9 via UAI (j) 34,236 75,000 154,248 137,343 122,058 133,963 656,848
11
Total est. additional from region
(L8 + L10)
34,236 75,000 154,248 137,343 122,058 133,963 656,848
12 Est. quantity from UAI (L3 + L6 + L11) 41,884 84,156 167,489 300,363 143,215 153,321 890,428
a. Ugandan production 1998-2003 based on USGS data (most recent Yearbook) and UNPE for 1998 (2001b: Table 1).
b. Based on either import or export data from ComTrade (no transaction was counted twice; different reporters for different years).
c. Uganda exports excess to production assumed to be of DRC origin because the only producing countries nearby were DRC, Rwanda and Burundi and
Ugandan trade in tin with these countries unlikely 1998-2003. Zimbabwe produced small quantities, but this would not be exported via Uganda.
d. Based on either import or export data from ComTrade (no transaction was counted twice; different reporters for different years).
e. See text for explanation of estimate that 95% of cassiterite production 1998-2003 was in non-Government area.
f. See text for explanation of estimate that 5% of cassiterite production in non-Government area was in UAI.
g. Based on ComTrade. Tanzania did not produce cassiterite 1998-2003. It was assumed its exports originated in the DRC.
h. We cannot know if Tanzanian exports originated in state-held or UAI areas, so assumed 50% may have originated in UAI.
i. Based on ComTrade. However, Congo-Brazzaville and its neighbours are not producers and it is too distant from DRC mines to be an export route,
especially 1998-2003. Assumed, therefore, that imports reported as from “Congo” were actually from the DRC.
j. Because I cannot know by what route Congo-Brazzaville exports left the DRC, I assumed 33% passed via Uganda.
124
Tin (Cassiterite): Value, USD
Table A4.5.4.2
*1998 and 2003 five mnths only
1998* 1999 2000 2001 2002 2003* Total
1 Est. quantity from UAI (a) 41,884 84,156 167,489 300,363 143,215 153,321 890,428
2 Est. price, USD per kg (b) 2.12 1.50 1.83 2.03 2.02 4.12
3 Total (L1 x L2) 88,905 126,347 306,847 609,436 289,024 632,341 2,052,900
4 To get 2020 USD multiply L3 by … (c) 1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD (L3 x L4) 142,248 197,102 463,339 895,870 419,084 891,601 3,009,244
a. From L12 in previous table.
b. Initial price estimate based on an average of all price observations for each year (1998-2003) available from ComTrade records for cassiterite
imports and exports involving East and Central Africa. This price was then reduced by 35% to better reflect probable price at points of opportunities
for exploitation in the DRC (always less than prices paid by international importers).
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
Tin (Cassiterite): Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.4.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kgs) 44,521 5% 845,907 95% 890,428
2 Base value of quantity (b) 150,462 5% 2,858,783 95% 3,009,245
3 Est. value of Theft (c) 7,523 5.0% 14,294 0.5% 21,817
4 Est. Fees & Licences (d) 7,147 5.0% 57,176 2.0% 64,323
5 Est. of taxes on Value (e) 28,588 20.0% 142,939 5.0% 171,527
6 Total est. value of damages $ 43,258 $ 214,409 $ 257,667
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
125
A4.5.5 Tungsten
Table A4.5.5.1
*1998 and 2003 five mnths only
1998* 1999 2000 2001 2002 2003* Total
Uganda
1 Est. production (a) 0 0 0 17,000 16,000 417 33,417
2 Est. exports (b) 0 82,237 70,600 116,885 60,000 12,500 342,222
3 Exports surplus to production; assume
from DRC (L1 - L2) (c)
0 82,237 70,600 99,885 44,000 12,083 308,805
D. R. Congo
4 Est. DRC exports (d) 0 0 0 0 0 11,667 11,667
5 Assume 95% of L4 from non-Govt area (e) 0 0 0 0 0 11,084 11,084
6 Assume 5% of L5 from UAI (f) 0 0 0 0 0 3,675 3,879
Other probable exporters of DRC tungsten
7 Est. exports from Kenya and Tanzania (g) 0 0 0 0 0 27,997 27,997
8 Assume 50% of L7 via UAI (h) 0 0 0 0 0 13,998 13,998
9 Est. exports from Congo-Brazzaville (i) 4,583 0 0 0 0 8,458 13,041
10 Assume 33% of L9 via UAI (j) 1,528 0 0 0 0 2,819 4,347
11 Total est. additional from region (L8 + L10) 1,528 0 0 0 0 16,817 18,345
12 Est. quantity from UAI (L3 + L6 + L11) 1,528 82,237 70,600 99,885 44,000 32,575 330,825
a. Ugandan production based on USGS data (most recent Yearbook)
b. Uganda exports based on ComTrade import data from “All” partners.
c. Uganda exports in excess of production were assumed to be of DRC origin because the only producing countries nearby were the DRC, Rwanda and
a small quantity from Burundi in 2003, but cross-border trade in tungsten with Rwanda and Burundi was unlikely 1998-2003.
d. Based on either import or export data from ComTrade (no transaction was counted twice; different reporters for different years).
e. See text for explanation of estimate that 95% of tungsten production 1998-2003 was in non-Government area.
f. See text for explanation of estimate that 5% of tungsten production in non-Government area was in UAI.
g. Based on ComTrade. Kenya and Tanzania did not produce tungsten 1998-2003. It was assumed their exports originated in the DRC.
h. We cannot know if Kenyan and Tanzanian exports originated in state-held or UAI areas, so assumed 50% may have originated in UAI.
i. Based on ComTrade. However, Congo-Brazzaville and its neighbours are not producers and it is too distant from DRC mines to be an export route,
especially 1998-2003. Assumed, therefore, that imports reported as from “Congo” were actually from the DRC.
j. Because I cannot know by what route Congo-Brazzaville exports left the DRC, I assumed 33% only passed via Uganda.
126
Tungsten: Value, USD
Table A4.5.5.2
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
1 Est. quantity from UAI (a) 1,528 82,237 70,600 99,885 44,000 32,575 330,825
2 Est. price, USD per kg (b) 2.48 2.00 3.49 3.34 2.87 3.66
3 Total (L1 x L2) 2,463 106,860 160,020 216,926 82,038 77,498 645,805
4 To get 2020 USD multiply L3 by … (c) 1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD (L3 x L4) 3,940 166,701 241,630 318,881 118,956 109,273 959,381
a. From L12 in previous table.
b. Initial price estimate based on an average of all price observations for each year (1998-2003) available from ComTrade records for tungsten imports
and exports involving East and Central Africa. This price was then reduced by 35% to better reflect probable price at points of opportunities for
exploitation in the DRC (always less than prices paid by international importers).
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
Tungsten: Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.5.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kgs) 16,541 5% 314,284 95% 330,825
2 Base value of quantity (b) 47,969 5% 911,411 95% 959,380
3 Est. value of Theft (c) 2,398 5.0% 4,557 0.5% 6,955
4 Est. Fees & Licences (d) 2,279 5.0% 18,228 2.0% 20,507
5 Est. of taxes on Value (e) 9,114 20.0% 45,571 5.0% 54,685
6 Total est. value of damages $ 13,791 $ 68,356 $ 82,147
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
127
A4.5.6 Timber
Table A4.5.6.1
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
D. R. Congo
1 Est. DRC exports (a) 20,023,126 8,825,283 50,622,170 10,144,661 17,493,130 14,555,837 121,664,207
2 Assume 80% of L1 from non-Govt area (b) 16,018,501 7,060,226 40,497,736 8,115,729 13,994,504 11,644,669 97,331,365
3 Assume 50% of non-Govt area is in UAI (c) 8,009,251 3,530,113 20,248,868 4,057,864 6,997,252 5,822,335 48,665,683
4 Est. informal exports from DRC to Uganda
and via Uganda to Kenya (d)
3,500,000 8,400,000 8,400,000 8,400,000 8,400,000 3,500,000 40,600,000
5 Uganda re-exports; assume DRC-origin (e) 0 0 3,620 96,327 0 3,750 103,697
6 Est. quantity from UAI (L3 + L4 + L5) 11,509,251 11,930,113 28,652,488 12,554,191 15,397,252 9,326,085 89,369,380
a. DRC exports in kilograms based on ComTrade import data from “All” partners. ‘Timber’ was defined using commodity code “HS 4407: Wood sawn
or chipped lengthwise, sliced, peeled, whether or not planed, sanded or finger jointed, or a thickness exceeding 6mm”. I.e., sawnwood. ComTrade
did not provide data for other wood which should be assumed have also left the DRC, such as industrial round logs, sawlogs or veneer logs, or other
wood products. ComTrade has four options for timber weight: kilograms, cubic metres, litres and ‘no quantity’. Where litres or no quantity were
recorded, approximate weight in kilograms was obtained by using an average price per kilogram from all price observations for the relevant year.
b. See text for explanation of estimate that 80% of timber production 1998-2003 was in non-Government area.
c. See text for explanation of estimate that 50% of timber production in non-Government area was in UAI.
d. See text for explanation, but estimate based on Umunay (2011) which was revised down to 8,400,000 (12% of Umunay’s estimate total) to reflect
reduced informal exports during 1998-2003.
e. ComTrade has some data for Ugandan re-exports, i.e., it imported timber and then re-exported it. It was assumed this was from the DRC.
128
Timber: Value, USD
Table A4.5.6.2
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
1 Est. quantity from UAI (a) 11,509,251 11,930,113 28,652,488 12,554,191 15,397,252 9,326,085 89,369,380
2 Est. price, USD per kg (b) 0.44 0.44 0.35 0.40 0.34 0.42
3 Total (L1 x L2) 5,012,279 5,195,564 9,897,388 5,082,654 5,247,164 3,888,311 34,323,360
4 To get 2020 USD multiply L3 by … (c) 1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD (L3 x L4) 8,019,646 8,105,080 14,945,056 7,471,501 7,608,387 5,482,519 51,632,189
a. From L6 in previous table.
b. Prices taken from International Tropical Timber Organization’s database: https://www.itto.int/biennal_review/?mode=searchdata. Searched for
Sawn wood (NC) > Exports Unit Value for 1998-2003. Given price was in cubic metres, so converted into USD per kg. This price was then reduced by
35% to better reflect probable price at points of opportunities for exploitation in the DRC (always less than prices paid by international importers).
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com.
Timber: Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.6.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kgs) 44,684,690 50% 44,684,690 50% 89,369,380
2 Base value of quantity (b) 25,816,095 50% 25,816,095 50% 51,632,190
3 Est. value of Theft (c) 516,322 2.0% 129,080 0.5% 645,402
4 Est. Fees & Licences (d) 252,998 1.0% 258,161 1.0% 511,159
5 Est. of taxes on Value (e) 2,023,982 8.0% 258,161 1.0% 2,282,143
6 Total est. value of damages $ 2,793,302 $ 645,402 $ 3,438,704
a. See text for explanation of Ituri and non-Ituri share of quantity and value.
b. From Total from L5 in previous table.
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
129
A4.5.7 Coffee
Table A4.5.7.1
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
D. R. Congo
1 Est. DRC exports (a) 11,782,835 24,293,751 20,965,544 11,095,656 4,965,936 2,279,140 75,382,862
2 Est. 80% of L1 is from non-Govt area (b) 9,426,268 19,435,001 16,772,435 8,876,525 3,972,749 1,823,312 60,306,290
3 Est. 50% of L2 is from UAI (c) 4,713,134 9,717,500 8,386,218 4,438,262 1,986,374 911,656 30,153,144
Uganda
4 Est. Ugandan exportable production (d) 82,612,500 180,390,000 180,060,000 188,280,000 173,010,000 65,225,000 869,577,500
5 Declared Ugandan imports from DRC (e) 3,078 3,078
6 Coffee available for export (L4 + L5) 82,612,500 180,390,000 180,060,000 188,280,000 173,010,000 65,228,078 869,580,578
7 Imports from Kenya, Rwanda, Burundi 203 1 223 427
8 Adjusted available coffee for export (L6
- L7) (f)
82,612,500 180,390,000 180,059,797 188,279,999 173,009,777 65,228,078 869,581,005
9 Recorded exports (g) 72,385,198 197,637,388 153,764,884 131,568,379 167,538,326 69,946,836 792,841,011
10 Available export coffee minus recorded
exports (L9 - L8) (h)
10,227,303 (17,247,388) 26,295,116 56,711,621 5,471,674 (4,721,836) 76,736,490
11 Adjusted unexplained surplus exports (i) 12,133,737 1,492,460 13,626,197
12 Est. quantity from UAI (L3 + L11) 4,713,134 21,851,237 8,386,218 4,438,262 1,986,374 2,404,116 43,779,341
a. DRC exports based on ComTrade import data from “All” partners.
b. See text for explanation of estimate that 80% of coffee production 1998-2003 was in non-Government area.
c. See text for explanation of estimate that 50% of coffee production in non-Government area was in UAI.
d. Uganda’s production of export coffee based on ICO data. See text for explanation.
e. Imports of DRC coffee could be re-exported so should be added to Uganda’s production (but few data anyway).
f. Coffee from Kenya, Rwanda or Burundi could be re-exported, so were subtracted from Uganda’s coffee available for export to ensure no confusion
with coffee originating from the DRC (but very few data anyway).
g. Uganda’s exports based on ComTrade import data from “All” partners.
h. Positive amounts indicate left over, un-exported, production. Negative amounts, shown in parentheses, mean Uganda exported more than it
produced raising the question as to where this coffee came from.
i. Green coffee beans can be stored for one year. In 1999 and 2003, exports were greater than production. These years were ‘discounted’ by an
amount equal to 50% of the previous year’s production surplus, in case some coffee was stored for 12 months before export.
130
Coffee: Value, USD
Table A4.5.7.2
*1998 and 2003 five months only
1998* 1999 2000 2001 2002 2003* Total
1 Est. quantity from UAI (a) 4,713,134 21,851,237 8,386,218 4,438,262 1,986,374 2,404,116 43,779,341
2 Est. price to DRC growers, USD per kg (b) 1.33 1.11 0.92 0.77 0.68 0.69
3 Total (L1 x L2) 6,268,468 24,254,873 7,715,320 3,417,462 1,350,735 1,658,840 44,665,698
4 To get 2020 USD multiply L3 by … (c) 1.60 1.56 1.51 1.47 1.45 1.41
5 Est. total value in 2020 USD (L3 x L4) 10,029,549 37,837,602 11,650,133 5,023,669 1,958,565 2,338,965 68,838,483
a. From L12 in previous table.
b. Prices based on International Coffee Organization’s “Prices to Growers” Historical Data on the Global Coffee Trade.
http://www.ico.org/new_historical.asp
c. Rates taken from US Inflation Calculator, based on US Government CPI data published on October 13, 2020, which uses US Labor Dept Bureau of
Labor Statistics data: https://www.usinflationcalculator.com
Coffee: Quantity and value distribution across Ituri and Non-Ituri, 2020 USD
Table A4.5.7.3 Ituri (a) % Non-Ituri (a) % Total UAI
1 Quantity (kgs) 13,133,802 30% 30,645,539 75% 43,779,341
2 Base value of quantity (b) 20,651,545 30% 48,186,938 70% 68,838,483
3 Est. value of Theft (c) 206,515 1.0% 0 0.0% 206,515
4 Est. Fees & Licences (d) 204,450 1.0% 240,935 0.0% 445,385
5 Est. of taxes on Value (e) 1,635,602 8.0% 481,869 1.0% 2,117,471
6 Total est. value of damages $ 2,046,568 $ 722,804 $ 2,769,372
a. See text for explanation of Ituri and non-Ituri share of quantity and value
b. From Total from L5 in previous table
c. See text for explanation of proxy ‘theft tax’.
d. See text for explanation of proxy ‘tax on fees and licences’.
e. See text for explanation of tax on value.
131
Appendix 4.6: Signature of Expert
This report has been prepared in accordance with the terms of reference set out by the International
Court of Justice by MICHAEL NEST on 19 December 2020:
Signed: