G
lobal
S
tudy on
UNITED NATIONS
New York, 2020
Global Study on
Firearms Trafficking
2020
UNITED NATIONS OFFICE ON DRUGS AND CRIME
Vienna
DISCLAIMER
This publication may be reproduced in whole or in part and in any form for educational or non-profit purposes
without special permission from the copyright holder, provided acknowledgement of the source is made.
Suggested citation: UNODC, Global Study on Firearms Trafficking, 2020
(United Nations publication, Sales No. E.20.IV.1).
The content of this publication does not necessarily reflect the views or policies of UNODC, Member States or
contributory organizations, and nor does it imply any endorsement.
This document has not been formally edited. The designations employed and the presentation of material in this
publication do not imply the expression of any opinion whatsoever on the part of the Secretariat of the United
Nations concerning the legal status of any country, territory, city or area, or of its authorities, or concerning the
delimitation of its frontiers or boundaries.
This document was produced with the financial assistance of the European Union. The views expressed herein
can in no way be taken to reflect the official opinion of the European Union.
© United Nations, March 2020. All rights reserved, worldwide.
Title: Global Study on Firearms Trafficking
Language: English
Sales no.: E.20.IV.1
ISBN: 978-92-1-130404-6
eISBN: 978-92-1-004969-6
3
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
CONTENTS
ABBREVIATIONS ............................................................................................................................. 5
EXECUTIVE SUMMARY ................................................................................................................... 7
INTRODUCTION ............................................................................................................................ 15
Seizures data to help understand firearms trafficking? ....................................................................................................... 16
Data sources and limitations
............................................................................................................................................. 17
CHAPTER 1 - THE PRIMARY EVIDENCE BASE: OVERVIEW OF SEIZURES ................................... 19
Chapter overview .............................................................................................................................................................. 19
How much is seized
.......................................................................................................................................................... 20
Types of firearms seized
..................................................................................................................................................... 25
Parts and components and ammunition
............................................................................................................................ 27
CHAPTER 2 - FROM SEIZURES TO TRAFFICKING ......................................................................... 31
Chapter overview .............................................................................................................................................................. 31
Why is it seized: legal justification and criminal context
................................................................................................... 32
Understanding the black market for firearms
.................................................................................................................... 36
SDG indicator 16.4.2 and tracing
..................................................................................................................................... 45
Licit manufacture
.............................................................................................................................................................. 50
CHAPTER 3 - THE NATURE OF FIREARMS TRAFFICKING ............................................................ 57
Chapter overview .............................................................................................................................................................. 57
Prices of firearms in illicit markets
..................................................................................................................................... 58
Modalities of trafficking
.................................................................................................................................................... 60
CHAPTER 4 - LINKS TO ORGANIZED CRIME, TERRORISM AND OTHER FORMS OF CRIME ...... 77
Chapter overview .............................................................................................................................................................. 77
Forms of crime linked to firearms
..................................................................................................................................... 78
CONCLUSIONS AND POLICY IMPLICATIONS ............................................................................... 85
ANNEX .......................................................................................................................................... 89
Africa ................................................................................................................................................................................ 89
Americas
........................................................................................................................................................................... 92
Europe
.............................................................................................................................................................................. 96
Asia and Oceania............................................................................................................................................................. 100
ANNEX - TYPES OF FIREARMS ................................................................................................... 102
ANNEX - GLOSSARY
................................................................................................................... 104
ANNEX - STATISTICAL ANNEX
................................................................................................... 107
Acknowledgements
The Global Study on Firearms Trafficking was prepared by the UNODC Crime Research Section, under the
supervision of Jean-Luc Lemahieu, Director of the Division for Policy Analysis and Public Affairs, and Angela Me,
Chief of the Research and Trend Analysis Branch. The UNODC Global Firearms Programme participated by
providing substantive input and advice under the supervision of John Brandolino, Director of the Division for Treaty
Affairs, and Loide Ayree, Chief of the Organized Crime and Illicit Trafficking Branch.
Core team
Analysis, drafting and study preparation
Kristiina Kangaspunta, Antoine Vella, Raggie Johansen, Simonetta Grassi, Leonardo Lara and Mareike Buettner
Database development and data management
Enrico Bisogno, Salomé Flores, Hernan Epstein, Francisco Guerreiro, Trung Anh Dang, Francesca Rosa,
Umidjon Rakhmonberdiev and Stefanie Mavrakou
Graphic design, layout and mapping
Anja Korenblik, Antero Keskinen, Lorenzo Vita, Suzanne Kunnen, Kristina Kuttnig and Federica Martinelli
Editor (Executive Summary)
Joseph Boyle
Technical inputs, review and comments
The Global Study on Firearms Trafficking benefited also from the expertise and invaluable contributions of other
UNODC colleagues in the Global Firearms Programme, namely: Diman Dimov, Paul Ianovici, Lucia Gomez
Consoli, Jacques Seckene Ndour and Silvia De Pedro Sanchez-Romero.
External inputs
This study also benefited greatly from data on customs seizures provided by the World Customs Organization. Data
from the Crime-Terror Collaboration Database (University of Massachusetts Lowell) was made available by Professor
Arie Perliger.
5
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
ABBREVIATIONS
ATF Bureau of Alcohol, Tobacco, Firearms
and Explosives (United States)
ARQ Annual Report Questionnaire
(UN Office on Drugs and Crime)
CTS United Nations Survey of Crime Trends and
Operations of Criminal Justice Systems
(UN Office on Drugs and Crime)
GFP Global Firearms Programme
(UN Office on Drugs and Crime)
IAFQ Illicit Arms Flows Questionnaire
IPFM Intimate partner or family member
MIAFI Monitoring Illicit Arms Flow Initiative
SALWs Small Arms and Light Weapons
Southern Europe
(excl. WB)
Southern Europe
(excluding Western Balkans)
UNODA United Nations Office for
Disarmament Affairs
UNODC United Nations Office on Drugs
and Crime
WCO World Customs Organization
7
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Firearms trafficking is a global
phenomenon with strong variations
between countries
UNODC has carried out its most comprehensive collec-
tion of data on firearms trafficking to date, gathering
details from survey responses and other sources in 81
countries for 2016-17. These data give a unique insight
into the patterns of trafficking globally, regionally and
nationally, providing a vital resource for law enforcement,
policymakers and public bodies seeking to reduce the
damage caused by the illicit circulation of firearms.
Based on these sources, a total of 550,000 firearms were
seized during each of 2016 and 2017. The data showed
wide variations among countries in terms of quantities
seized, which ranged from less than 10 to more than
300,000. The nature of seizures reported also varied
dramatically.
The real global figure for seizures is much higher than
550,000, as some of the countries covered by this study
underreported their seizures for administrative reasons,
and the quality of data varied significantly between coun-
tries. Also, many countries have not provided any infor-
mation, including some of the world’s most populous
nations. Nonetheless, the coverage for this attempt at
global data collection is good and likely to improve in the
coming years, as data collection is streamlined and embed-
ded in national institutions.
Pistols emerge as most seized
firearms globally
Pistols are the world’s most seized type of firearm. How-
ever, this pattern is driven to a large extent by the Ameri-
cas, the region that reported the most seized firearms
overall. Pistols constituted more than 50 per cent of the
total firearms seized in the region during the reporting
period.
In Africa and Asia, shotguns were the most prominent
type. Rifles were the main type of firearm seized in Oce-
ania, and in Europe the distribution was more equal
between pistols, rifles and shotguns.
EXECUTIVE
SUMMARY
8
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 1
Typical distribution* of reported seized arms, by type, 2016-17
* Simple average based on data for 81 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms or
small arms and light weapons (SALWs).
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Source: UNODC IAFQ and other official sources.
However, many countries in Africa and Asia appear to
have a lower capacity to intercept and report trafficked
firearms, which may lead to underreporting of some types
of firearms. Moreover, the total figures reported by coun-
tries include seizures which are not directly connected to
trafficking. Based on customs seizures at borders, rifles
emerge at par with pistols. This suggests that firearms such
as rifles may play a bigger role in global trafficking pat-
terns than what is reflected in the currently available data.
Looking more closely, links emerge between trafficking
patterns and broader regional contexts. For example, coun-
tries with higher levels of violent deaths and homicide,
particularly in Africa and Latin America and the Carib-
bean, tend to seize a higher percentage of firearms con-
nected to violent crime. Similarly, in countries with higher
levels of drug trafficking, more arms are seized linked to
that activity.
FIG. 2
Total number of arms seized, by country and type of coverage, 2016-17 (average)
¹ For Canada, Ghana and Paraguay data were available for 2016 only.
² For Cameroon, Guinea and Kyrgyzstan data were available for 2017 only.
Sources: UNODC IAFQ and other official sources.
1
10
100
1,000
10,000
100,000
1,000,000
Australia
Canada¹
ArgenƟna
Russian FederaƟon
Spain
Kenya
Mexico
Greece
Italy
Guatemala
Peru
El Salvador
Ecuador
Sudan
Slovakia
Denmark
Luxembourg
Tunisia
Japan
Jamaica
Albania
Lithuania
Central African Republic
Cote d'Ivoire
Suriname
Morocco
Burundi
Togo
Qatar
Paraguay¹
Saint Lucia
Liechtenstein
AnƟgua and Barbuda
Grenada
United States of America
Costa Rica
Portugal
Honduras
Brazil
Ukraine
CroaƟa
Lebanon
Dominican Republic
United Kingdom
Algeria
Burkina Faso
Hungary
Libya
Azerbaijan
Guyana
Cameroon²
Romania
Republic of Moldova
Nepal
Philippines
Gh ana¹
Colombia
Angola
Belarus
Chile
Kazakhstan
Netherlands
Uruguay
Belgium
Serbia
Panama
Sweden
Norway
Montenegro
Bahamas
Slovenia
Myanmar
Cabo Verde
North Macedonia
Bolivia (PlurinaƟonal State of)
Kyrgyzstan²
Guinea²
Botswana
Tajikistan
MauriƟus
Cuba
FULL COVERAGE INCOMPLETE COVERAGE NO INFORMATION ON COVERAGE/UNCLEAR
Number of arms (logaithmic scale)
Does not include administraƟve seizures
EŽŝŶĨŽƌŵĂƟŽŶŽŶĂĚŵŝŶŝƐƚƌĂƟǀĞƐĞŝnjƵƌĞƐͬƵŶĐůĞĂƌ
Includes administraƟve seizures
Other undistinguished weapons
(unspecified/aggregate)**
Pneumatic, blank-firing
and gas weapons***
Weapons or related
items other than
firearms/SALWs,
including:***
Unclassified/
unknown
Pistols,
39%
Rifles,
18%
Revolvers,
14%
Submachine
guns, 3%
Machine
guns, 1%
Other firearms
Other SALWs
Shotguns,
25%
Firearms/
SALWs,
72%
9
Introduction
FIG. 3
Typical distributions* of seized fire-
arms/SALWs by type, according to
region, 2016-17
* Simple averages, adjusted for any firearms/SALWs which could not be
classified and quantified into the respective category and weapons other
than firearms/SALWS.
Source: UNODC IAFQ and other official sources.
MAP 1
Main transnational firearms trafficking flows (as defined by routes of seized firearms),
2016-17
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
Transnational trafficking exists
alongside domestic diversion and
illicit manufacture
Most firearms seizures are made within national territories;
seizures at borders on average account for less than 10 per
cent of all interceptions. Seized weapons are overwhelm-
ingly manufactured outside the country of seizure, but it
is likely that the sourcing of firearms found in illicit mar-
kets has an important domestic component, such as fire-
arms diverted from licit channels in the country of
seizure.
There is often little connection to the country of manu-
facture – a significant proportion of arms seized on incom-
ing shipments have not been transported directly from
where they were produced. This suggests that vulnerability
to firearms trafficking is mostly to be found in countries
where firearms are diverted from legal holdings rather than
where they are manufactured.
Northern America is the principal subregion of departure
for seized firearms, according to the available data. On a
smaller scale, Europe and Western Asia are also major
28%
52%
34%
35%
7%
22%
8%
16%
27%
71%
24%
10%
11%
38%
14%
37%
22%
20%
8%
0% 20% 40% 60% 80% 100%
Africa
(18 countries)
Americas
(26 countries)
Asia
(10 countries)
Europe
(26 countries)
Oceania
(1 country)
Pistols Rifles Revolvers
Shotguns Submachine guns Machine guns
Subregions
Eastern Europe
Northern AfricaCentral America
Caribbean
South America
Northern America
Northern and Western Europe Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
sĞƌLJƐŵĂůůǀŽůƵŵĞŇŽǁ
Small
Medium
Large
Southern Europe (excl. WB)
10
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
departure points for illicit flows. However, illicit flows
within subregions often account for significant propor-
tions of firearms entering the market, notably in South
America, Northern and Western Europe and Western Asia.
Transnational trafficking
is concentrated within continents
Based on the routes of seized firearms, transnational traf-
ficking flows seem to be mostly concentrated within con-
tinents. Northern America plays a significant role as
departing subregion for other subregions, particularly
South and Central America, as well as Western Asia.
Northern America, Europe and Western Asia together
accounted for almost all departure points of trafficking in
2016-17. In contrast with other parts of the world, outgo-
ing flows from countries in Europe were predominantly
intended for trafficking across continents.
FIG. 4
Distribution of subregions identified as origin of
incoming illicit flowsª (inner circle) and
manufactureª (external circle) for seizures made
in South America, 2016-17
*Excluding Western Balkans.
ªThe largest share of transnational illicit flows affecting countries in South America occurs
between countries within South America. Similarly, countries in South America itself
account for the largest share of manufacture of weapons seized in South America
(including weapons seized in the country of manufacture). These shares are not shown in
the above figure.
Note: The shares of flows and of manufacture are based on different kinds of data which
require different methodologies. Therefore, the comparison should be made with caution.
In both cases, the calculations adjust for the share which is not classified or reported as
unknown.
Source: UNODC IAFQ and other official sources.
Northern
America
89%
Northern
and
Western
Europe
7%
Southern
Europe*
2%
Western
Asia
2%
Eastern and
South-
Eastern Asia
Eastern
Europe
2%
Northern
America
17%
Northern
and
Western
Europe
10%
Southern
Europe*
66%
Western
Asia
5%
Western
Balkans
Flows
Manufacture
Central and South America together with Western Asia
accounted for more than 80 per cent of trafficking desti-
nations. These main destination areas are known for high
levels of criminal violence or conflict and show the links
between firearms trafficking and violent deaths.
Manufacturing country often
unconnected to illicit flow
Illicit firearms flows are complex and do not necessarily
follow licit flows. The country of manufacture of firearms
and the country where diversions (when firearms leave the
licit circuit and enter the illicit one) and seizures take place
often do not overlap. This is clearly seen in the contrast
between countries that are identified either as manufactur-
ing countries or as the point where the illicit flow starts.
While Europe emerges as the main manufacturing region
in seizures made across the world, the most prominent
subregion of illicit origin is Northern America. Firearms
are durable goods and their circulation before and after
diversion to the black market often involves several
transfers.
Cross-border seizures are most
common at land crossings, but sea
shipments are bigger
Traffickers tend to use sea transport for large shipments.
Cases of seizures from vessels involved more than five times
the number of firearms typically intercepted from other
types of transportation. This suggests that law enforcement
could get a better return on their investment if they
focused on transportation by sea.
However, trafficking by land remains the most common
type of cross-border case, accounting for roughly two-
thirds of the total. Interceptions from vessels accounted
for only around 6 per cent of all customs cases, but 33 per
cent of the total number of firearms seized by customs.
Size of seizure case can reflect the
intended use of the firearm
It appears that the vast majority of seizure cases entail law
enforcement officers intercepting a single firearm or a very
small number, according to available data. However, in
terms of quantities of firearms seized, the big cases may
account for a significant share. For example, among cus-
toms seizures carried out at national borders, roughly three
quarters of cases involved one firearm, but around one half
of firearms were seized in instances that involved 18 or
more firearms.
Seizures of small consignments of one or two firearms may
be linked to individual use, where the firearm is taken
from the end user. These cases most commonly involved
handguns. For example, more than 80 per cent of revolv-
ers were captured in cases involving just a single firearm.
11
Introduction
This type of case often entails a violation of possession
regulations, but can also involve strategic “ant trafficking”,
whereby many people transport weapons in small consign-
ments to meet large-scale demand and reduce the risk of
disruption by law enforcement. This type of trafficking
does not fully explain global arms trafficking, but there is
evidence
1
that it is utilized to transport firearms from the
United States to Mexico.
At the other end of this continuum, countries reported
large seizures that seemed to respond to big instances of
demand, for example conflict. Seizures of rifles, shotguns
and pistols suggest that firearms of these types can be traf-
ficked in consignments of hundreds and thousands. High-
powered firearms such as machine guns and submachine
guns are not often captured, but these rare seizure events
generally involve larger quantities. Around one fifth of all
customs seizure cases that included machine guns and sub-
machine guns involved such firearms in batches of four or
more; on the other hand, such hauls of revolvers, for exam-
ple, were exceptional.
These big seizure cases were far more likely than smaller
consignments to be linked to firearms trafficking offences.
1 See Section on “Ant trafficking” in Chapter 3.
Illicit sellers can compete with licit
markets on price
The cost of buying firearms in the illicit market is usually
significantly higher than in the licit sphere, reflecting the
extra risks and profiteering involved in the black market.
However, there are exceptions. In Latin America and the
Caribbean, for example, illicit prices were lower than the
licit price for handguns. This suggests ample availability
of handguns in the illicit market.
Price data from Europe point to the Western Balkans as a
potential illicit source of firearms, notably assault rifles.
Such weapons were far less expensive in the Western Bal-
kans than in the rest of Europe.
Traffickers supply those seeking to
assemble or convert weapons
Seizures of parts and components are relatively rare com-
pared with seizures of firearms: on average, only 5 per cent
of the number of firearms. However, a different picture
emerges in countries where firearms in non-factory condi-
tion were seized in relatively high numbers (including arti-
sanal or craft production), particularly in Africa and parts
of Asia. Countries in these areas tend to report relatively
high levels of seizures of parts and components.
Illicit manufacturing, conversion, reactivation and assem-
bly of firearms is also present in some European countries,
Border
Border
Border
Border
d
Borde
Border
Border
Border
Border
Border
Border
Land
TRAFFICKING
Small quantities
in many different
vehicles
Ant
TRAFFICKING
Many actors/traffickers
carrying small amounts
Sea
TRAFFICKING
Big quantities are
smuggled by boats
CROSS-BORDER FIREARMS TRAFFICKING METHODS
12
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
possibly reflecting national control mechanisms that make
firearms hard to access legally. Modification of weapons
may also serve the needs of the criminals to use smaller or
more powerful weapons.
Firearms tracing remains insufficient
and countries risk missing SDG target
Countries are committed to “significantly reduce illicit
financial and arms flows” under SDG Target 16.4. One
of the indicators for this target is that the proportion of
firearms traced to their illicit origins must be measured.
Data from 14 countries in 2016-2017 suggest that, on
average, this illicit origin or illicit context was established
for just 28 per cent of the relevant category of firearms.
The countries that registered very high success rates in
tracing firearms reported relatively low seizure levels, per-
haps because tracing requires a lot of resources. On the
other hand, some countries with high levels of seizures
registered a low success rate, which may also be linked to
firearms seized in connection with less serious offences
and perhaps not prioritised in tracing.
Criminal justice responses tend
to underplay the significance of
firearms trafficking
Countries on average seized around two-thirds of firearms
on the grounds of illicit possession, according to the legal
justifications given by national authorities. Trafficking was,
on average, named as the legal justification in only around
9 per cent of cases.
However, it is likely that the offence of illicit possession
provides an easier and quicker way for law enforcement
to justify stopping shipments and seizing firearms, and
trafficking emerges as the actual offence only after further
investigations – this is known as the “criminal context” of
the seizure.
When the criminal context is factored in, the proportion
of seized weapons that could be connected to trafficking
more than doubles to roughly 19 per cent. The real pro-
portion is likely to be even higher, once potential under-
reporting of firearms trafficking is considered.
The evidence suggests that the criminal justice system
focuses on firearms trafficking only in a relatively small
percentage of the cases where it would be warranted –
meaning firearms trafficking is a largely hidden phenom-
enon, only part of which comes to the surface.
Violent crime and drug trafficking
are frequently linked with seizures
Other than arms offences, the criminal conduct most com-
monly linked with firearms seizures was violent crime,
particularly in Latin America and Africa. In Europe, drug
trafficking was the biggest category.
On average, a larger share of firearms is seized in the con-
text of violent crime in countries with high homicide rates.
The same link is shown with drug seizures. In addition,
drugs emerge as the most common commodity intercepted
in the same seizures as firearms, followed by counterfeit
goods, cultural property and natural resources
Unpicking the link between firearms trafficking and the
broader criminal context in which seizures occur is diffi-
cult. Some exceptionally large individual seizure cases
appear to be connected to areas with recent or ongoing
FIG. 5
Breakdowns of customs seizures, by
size* of seizure case, 2016-17
*
The size of a seizure case refers to the number of firearms seized in
that particular instance.
a
A case is considered “small” if between 2 and 5 firearms were seized
in that particular instance.
b
A case is considered “medium” if between 6 and 10 firearms were
seized in that particular instance.
c
A case is considered “large” if between 11 and 17 firearms were
seized in that particular instance.
d
A case is considered “exceptionally large” if at least 18 firearms were
seized in that particular instance.
Source: World Customs Organization.
Single
firearm,
74%
Small
cases
a
,
20%
Medium
cases
b
, 2%
Large
cases
c
, 1%
Exceponally
large cases
d
, 2%
Breakdown of seizure cases, by size*
Single
firearm,
24%
Small
cases
a
,
17%
Medium
cases
b
, 6%
Large cases
c
,
4%
Exceponally
large cases
d
,
49%
Breakdown of firearms seized,
by size* of seizure case
13
Introduction
conflicts, or countries with high levels of violence linked
to organized crime. But several large hauls were reported
that were apparently unconnected to conflict areas or
organized crime.
Some countries may face specific problems related to con-
flict. Weapons feed conflict while it is going on, then
stockpiles can proliferate in the aftermath, causing multi-
ple difficulties for the authorities.
More than 50 per cent of homicides
globally are carried out using
firearms
Overall, more than 50 per cent of homicides globally each
year are carried out with a firearm. The availability of fire-
arms is linked to the homicide rate: a rise in the rate of
firearms possession in a country often goes together with
an increase in the homicide rate.
However, the significance of firearms varies depending on
the context of the homicide. For example, firearms are by
far the most significant method in homicides related to
gangs or organized crime, but far less prominent when the
homicide involves intimate partners and family
members.
FIG. 6
Firearms seized in typical
a
customs seizure cases, by size
b
of seizure case (excluding cases
with atypical very large number of firearms), 2016-17
a
Isolated cases of a very large size were excluded; these were defined as cases with a size above the 98th percentile for the corresponding specific type
of firearm.
b
Size is measured as the number of firearms of the given specific type seized in the
corresponding case.
Source: World Customs Organization.
There is also a gender factor involved in the use of firearms
in homicide. Most homicide victims and perpetrators
globally are men, and this tends to be even more pro-
nounced among firearms homicides. When considering
homicides of intimate partners and family members, in
which most victims are women, men were more likely than
women to use a firearm when killing their female partners,
while women were more likely than men to resort to a
sharp object.
Overall, seizure data tended to suggest a relationship
between the level of interception and the rate of homicide.
Countries with higher levels of firearms seizures relative
to firearms homicides tend to have lower levels of homi-
cide, which might reflect an established and strong rule of
law situation.
1 revolver
1 pistol
1 shotgun
1 rifle
1 machine gun/
submachine gun
Small batches
Small batches
Small batches
Small batches
Small batches
Medium+
Medium+
Medium+
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Revolvers
Pistols
Shotguns
Rifles
Machine guns and
submachine guns
Seized on its own
Seized in small batches (2-5)
Seized in batches of 6 or more
15
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
INTRODUCTION
Firearms trafficking affects all parts of the world and
impacts on society in multiple ways. It is a major concern
in the context of human security, and it is central in law
enforcement efforts and activities. Firearms are instrumen-
tal in much violence, particularly homicide; they are often
used by organized criminals and they support operations
related to armed conflicts and terrorism. The most
common form of weapons trafficking involves small arms
and light weapons; however, the character of this traffick-
ing can vary significantly in different geographical contexts
and in relation to different weapons.
Firearms are usually manufactured for legal markets by
licenced manufacturers. They can, however, be diverted
into illegal markets at any point in their life cycle. The
action of law enforcement aims at stopping their illegal
movement. In many parts of the world, firearms are easily
available for those who can afford them. Firearms can cir-
culate in legal and illegal markets for a long time and
because they are durable commodities, they can be easily
reused and resold. This durability presents challenges to
prevention and control activities. For this report, data from
firearms seizures are used to have a closer look at firearms
trafficking. Building on the 2015 UNODC Study on Fire-
arms, this report analyses the flows of firearms trafficking,
the types of firearms that are trafficked, how this traffick-
ing is conducted and how it is related to other types of
crime. The report is not aiming at estimating the value of
the illegal market because the available data is too sporadic
to support such an estimation.
The complex nature of firearms concerns is also reflected
in the international legal framework. There is a wide vari-
ety of international and regional instruments addressing
firearms and their trafficking. The overall framework of
this report is based on the Protocol against the Illicit Man-
ufacturing of and Trafficking in Firearms, their Parts and
Components and Ammunition (‘the Firearms Protocol’)
supplementing the United Nations Convention against
Transnational Organized Crime which was adopted by the
General Assembly in 2001
1
. UNODC is the guardian of
the Convention and its Protocols. The Firearms Protocol,
which had 118 Parties as of October 2019, addresses the
illicit manufacturing of and trafficking in firearms from
the criminal justice angle, with a view to provide Member
States with measures to address the transnational nature
of the phenomenon and its links to organized and other
serious crime. Many other instruments introduce a com-
plementary approach to firearms trafficking from disarma-
ment, trade or development perspectives.
2
1 GA resolution 55/255 of 31 May 2001.
2 See a list of relevant instrument and documents in https://www.unodc.
org/unodc/en/firearms-protocol/international-legal-framework.html.
16
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 7
Circumstances of firearms seizures
In the context of the Conference of the Parties to the UN
Convention against Transnational Organized Crime,
Member States have requested UNODC to collect and
analyse quantitative and qualitative information and suit-
ably disaggregated data on trafficking in firearms, their
parts and components and ammunition
3
. In addition,
within the framework of the 2030 Agenda on Sustainable
Development, UNODC is monitoring the global devel-
opments related to the indicator 16.4.2 (“Proportion of
seized, found or surrendered arms whose illicit origin or
context has been traced or established by a competent
authority in line with international instruments”).
UNODC also serves as the international agency co-cus-
todian of this indicator together with the Office of Disar-
mament Affairs (ODA).
Seizures data to help understand
firearms trafficking?
In the United Nations Convention against Transnational
Organized Crime, “seizure (or freezing)” is defined as
“temporarily prohibiting the transfer, conversion, disposi-
tion or movement of property or temporarily assuming
custody or control of property on the basis of an order
issued by a court or other competent authority” (article
2f). Seizure can precede final confiscation or forfeiture,
defined as “permanent deprivation of property by order
of a court or other competent authority” (article 2g).
Criteria for seizing firearms can vary considerably among
different countries. The Firearms Protocol requires states
to seize firearms which are illicitly trafficked or manufac-
tured and to criminalize these offences. National legisla-
3 Conference of the Parties to the UN Convention against Transna-
tional Organized Crime, Ninth Session, Resolution 9/2, paragraph 32.
tion and regulations often go beyond the provisions of the
Protocol and firearms can also be seized for other reasons
such as having been used in criminal activities. Seizures
can also be based on administrative violations such as a
lack of a valid licence for possession or failure to comply
with storage conditions.
Seizures are a reflection of a complex reality and of differ-
ent overlapping facets of the phenomenon of illicit traf-
ficking
4
of firearms. Seizures may be related to criminal
activity but some seizures may also arise out of infringe-
ments of an administrative nature. Whether a particular
situation or conduct constitutes an administrative or a
criminal offence will depend on the national legislation,
but administrative offences would typically relate to
expired licences, improper storage, inadequate mainte-
nance, carrying or transporting a firearm in violation of
applicable restrictions, etc. There are also seizures which
happen as a preventive measure, typically in cases of
domestic violence or threats of violence - even if no crime
has been committed with the firearm.
Bearing in mind that firearms are durable goods which
can last for decades, firearms may be detected in criminal
settings long after they entered the black market (through
diversion or illicit manufacture). Moreover, some criminal
links which give rise to the seizure of a firearm may occur
independently of whether the firearm had been previously
trafficked or not. For example, a firearm can be used in
the commission of a crime, and therefore seized, whether
it was legally held or not. Sometimes a firearm is seized in
the context of crimes or investigations which are unrelated
4 The UN Firearms Protocol uses the term “illicit trafficking”. For edi-
torial reasons, this term will be shortened to “trafficking” throughout
this report.
CIRCUMSTANCES OF FIREARMS SEIZURES
CRIMINAL OFFENCES
ADMINISTRATIVE INFRINGEMENTS
PREVENTIVE MEASURES
(e.g. from legŝƟmate owner in context of
domesƟĐ violence)
ILLICIT SUPPLY CHAIN (“TRAFFICKING”)
Firearms exchanging hands, or crossing borders
at moment of deteĐƟŽn
HELD BY “END USER”
(no indicaƟŽn of transfers/cross-border
movement/intenƟon to pass on Įrearm at
moment of deteĐƟŽn)
• ClandesƟne shipments
• False/misleading declaraƟŽns
• Fake/decoy recipients
• Missing authorizaƟon or
licences
FIREARMS ILLICITLY EXCHANGING
HANDS DOMESTICALLY
• Illicit sales/transfers of legally
held Įrearms, including by/from
individuals and licensed dealers
• TheŌof civilian Įrearms
(usually small scale)
• Provision of/distribuƟon of/trade
in Įrearms of illicit origin,
including renƟng and pooling
• Diversion from law
enforcement/military holdings
(potenƟĂůůy large scale)
ILLICITLY MANUFACTURED/
ILLICIT CONDITION
• Rudimentary arms
• ModiĮĐaƟon
• Conversion
• ReaĐƟvaƟon
• Erased/altered markings
• Assembled from parts and
components
• Arms manufactured arƟsanally
without authorizaƟon
ILLICIT USE
• Homicide
• Bodily harm
• Threats and coercion
• Danger to public safety
• Armed robbery
ILLICIT CONTEXT
Independent invesƟgaƟons
into drug traĸĐking, traĸĐking
in persons, smuggling, other
organized crime, terrorism, etc.
“PURE” ILLICIT POSSESSION
Unauthorized possession, no other
apparent criminal links and unclear
origin
TRANSNATIONAL TRAFFICKING
AND DIVERSION
17
Introduction
to arms offences, such as organized crime or drug traffick-
ing; in such cases, trafficking may or may not have hap-
pened prior to the seizure. A firearm can also be seized
because of its condition, such as the case of altered or
deleted markings or illegal modifications, or because it was
illegally manufactured, including conversion from other
weapons, illegal assembly or illicitly manufactured in arti-
sanal settings. Once more, such firearms may or may not
have been trafficked prior to seizure. There may also be
cases of a criminal nature in which a firearm seized from
its holder is of unclear origin but there are no evident
criminal links other than unauthorized possession.
The cases described above typically occur when a firearm
is seized from its “end user”, that is, a person or group of
persons who held the firearm for their own purposes and
use (lawful or not). This leaves the case of firearms which
were detected in circumstances which suggest illegal trans-
fers or movement at the moment of interception. Even
among such firearms, such transfers or physical movement
may be internal to a country, or transnational. Such sei-
zures are those which reflect most directly the nature of
illicit flows of firearms. Seizures from cross-border ship-
ments, such as those made by Customs, are of particular
interest.
Data sources and limitations
The report is based on data collected from Member States
through the Illicit Arms Flow Questionnaire (IAFQ),
5
which was developed in cooperation with national and
international experts, relevant international and regional
organizations, and non-governmental organizations and
research entities, in order to collect seizure data in both
aggregate form and on a case-by-case basis. The question-
naire collected data on several aspects of firearms traffick-
ing, such as trafficking routes, criminal context, tracing
and the criminal justice system response to this crime. A
total of 80 responses to the questionnaire were received.
In addition, complementary sources of data were used,
particularly national reports on the Implementation of the
Programme of Action on Small Arms and Light Weapons
and the International Tracing Instrument (collected and
shared by the UN Office for Disarmament Affairs) and
seizure data from the World Customs Organization
(WCO). Most of the data cover the years 2016-2017.
5 Questionnaire on Illicit Arms Flows. Note by the Secretariat. Sixth
Session of the open-ended intergovernmental working group on
firearms established by the Conference of the Parties to the United
Nations Convention against Transnational Organized Crime, Vienna,
2-3 May 2018. CTOC/COP/WG.6/2018/CRP.2.
MAP 2
Responses to the UN-IAFQ (2018 data collection exercise on firearms) and other countries/
territories with partial or supplementary data from other sources
* Includes responses from 6 countries limited to metadata and qualitative data only.
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
Dashed lines represent undetermined boundaries. Dotted line represents approximately the Line of Control in Jammu and Kashmir agreed upon by India
and Pakistan. The final status of Jammu and Kashmir has not yet been agreed upon by the parties. The final boundary between the Sudan and South
Sudan has not yet been determined.A dispute exists between the Governments of Argentina and the United Kingdom of Great Britain and Northern
Ireland concerning sovereignty over the Falkland Islands (Malvinas).
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Responses to the UN-IAFQ (Illicit Arms Flows Questionnaire)*
Countries or territories with partial data for IAFQ indicators (2016/17) from other sources
18
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
These sources were used in particular to generate quanti-
tative data for some of the indicators covered by the Illicit
Arms Flow Questionnaire, yielding data for a total of 107
countries and territories for at least one of the years 2016
or 2017.
There are some limitations concerning the use of seizure
data in the context of firearms trafficking. Seizures can be
made for reasons other than firearms trafficking per se.
Firearms may be seized because they were used in the con-
text of criminal activities such as drug trafficking or homi-
cide. In addition, as for most such data related to crime,
seizure data reflect both the level of trafficking and the
effort and capacity of law enforcement to implement the
regulatory mechanisms which may vary considerably
between countries. Triangulating seizures data with other
information and analysing them across countries help to
use seizures data to inform patterns and dynamics of traf-
ficking but seizures alone do not describe the level of traf-
ficking. Legal definitions and regulatory frameworks can
differ greatly from one country to the next, making coun-
try-level comparisons difficult and fraught with risks of
comparing different phenomena. In addition to these gen-
eral limitations, some specific limitations concerning the
data collection for this report need to be noted. The data
collection could not cover all countries in the world and
even for those countries that did respond, the richness and
quality of the responses varied. While some parts of the
questionnaire resulted in good quality data, other parts
were left empty or only scarce data were submitted. These
limitations should be kept in mind when reading the
report.
19
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
CHAPTER 1
THE PRIMARY EVIDENCE BASE:
OVERVIEW OF SEIZURES
Chapter overview
This first chapter provides an overview of the available
data on firearms seizures across the world, focusing par-
ticularly on the magnitude and location of seizures, as well
as the types of firearms that are most frequently seized. It
also analyses the typical sizes of seizure events according
to type of firearm.
Data on total arms seized were available for 81 countries
(including IAFQ responses as well as other official sources),
amounting to roughly 550,000 arms in each of 2016 and
2017. This absolute figure is however dominated by the
Americas, due to very large quantities seized by some coun-
tries in this region. This region was also relatively well
represented in terms of the number of countries with avail-
able data. On the other hand, there were significant gaps
in data availability in Africa, Asia and Europe, including
missing data from relatively large countries. Moreover, in
some cases there were clear shortfalls arising from incom-
plete coverage within a single country. For these reasons
the analysis often focuses on averages representative of a
typical country rather than global totals.
Based on the available seizure data, pistols are the most
frequently seized firearm type. This is driven to a large
extent, however, by the pattern in the Americas. The
regional patterns vary considerably. While in the Americas,
the main types of seized firearms are handguns (pistols and
revolvers), in Africa and Asia, shotguns are most frequently
seized. In Europe, seizures are relatively evenly distributed
between pistols, rifles and shotguns, whereas in Oceania,
rifles appear to be the most seized type, although this is
based on data from only one country, Australia.
Considering the regional heterogeneity, the seizures pat
-
terns may also reflect trafficking patterns, with pistols and
revolvers being the most trafficked firearms in the Ameri-
cas and – to a lesser extent - in Europe. It is not clear,
however, how the global pattern of seized firearms reflects
global trafficking as there may be less capacity in countries
in Africa and Asia to intercept and report illicit shipments
of firearms, as indicated by the considerable data gaps in
these regions. It could be that the type of firearm domi-
nating seizures in Africa and Asia – shotguns – is more
significant in global trafficking flows than what can be
gleaned from available seizure data.
Firearms seizure events vary in size, although the vast
majority of seizures involve one or a few firearms. The
average number of firearms seized per case is 1.4 but in
terms of number of seized firearms, about half were seized
in large quantities. Most seizure events were small, but a
few cases with large quantities of firearms – notably rifles,
shotguns or pistols – were seized in a single instance. This
suggests that firearms can also be trafficked in large quan-
tities. While small scale trafficking is possible and present
20
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
in some regions (see section on ant trafficking in Chapter
3), it is likely that very small seizure cases (such as a single
firearm on its own) are frequently connected to other sce-
narios that might give rise to seizures, such as administra-
tive violations, illicit possession, perpetration or threats of
violence, endangering public safety, unauthorized manu-
facture, or firearms held by criminals which come to light
in the context of investigations unrelated to firearms (inde-
pendently of whether they had been trafficked at an earlier
point in time).
Data on seizures reflect both the patterns of the actual
firearms trafficking that is taking place and the efforts of
national law enforcement and other relevant authorities
in tackling it. As such, the data cannot be taken at face
value to accurately depict firearms trafficking flows. More-
over, the data should be interpreted cautiously, particularly
in terms of cross-country comparability. Some countries
have provided partial data that do not include seizures
made in certain geographical parts or by some government
entities. Countries have also made different choices regard-
ing the reporting of administrative and criminal seizures;
some reported separately, some jointly, and some were not
in a position to distinguish. These differences should be
taken into account in comparative analyses.
How much is seized
Seizures of firearms occur when agencies such as police,
customs or other law enforcement entities detect firearms
in circumstances which indicate or suggest a violation of
the law or applicable regulations. Seizures made by such
agencies may occur in the course of, inter alia, routine
inspections (including at border control points), targeted
operations, investigations into crimes and responses to
reported violations.
Thus, seizures reflect the primary point of contact between
the illicit phenomenon and the efforts of the state to coun-
ter it. Seizure data are influenced by, and therefore reflect,
Advantages and disadvantages of other data sources related to seizures
The evidence provided in this study relies on seizures
data together with other information that help to
interpret them in understanding firearms trafficking.
There are other data not considered in the study
linked to Government responses which could poten-
tially be used with or as an alternative to seizures, but
they are either not available systematically across
countries or they carry similar challenges to seizures.
Confiscations. While seizures represent a temporary
measure that can occur on various legal grounds and
usually signify the starting point of an investigation, a
confiscation is a permanent deprivation of property
that occurs at the end of a judicial proceeding when
factual circumstances are established.* While confis-
cated firearms could provide more reliable informa-
tion on trafficking of firearms, there are several
limitations. The representativeness of confiscations
data leans more than seizures towards measuring the
effectiveness of the criminal justice system rather
than illicit flows. Judicial confiscation orders also may
not exclusively relate to illicitly sourced and trafficked
firearms. Furthermore, data on confiscations are not
as readily available as data on seizures, as judicial
proceedings may take a long time and the informa-
tion may remain in individual files with no aggrega-
tion at national or sub-national level.
Circumstances of the seizure. Data on the circum-
stances of a seizure, such as its suspected criminal
context and its geographical occurrence, represent
precious complementary information that help con-
textualize and interpret seizure data for the analysis
of illicit flows, and questions on such aspects have
hence been included in the Illicit Arms Flows Ques-
tionnaire. Unfortunately, in many countries data at
such a level of disaggregation is not collected by
frontline officers. The challenge relates to the differ-
ent sources of information for seizures and their cir-
cumstances. Seizure data are typically compiled by
law enforcement authorities, while circumstances
may be recorded or revised by other criminal justice
institutions (prosecution and conviction institutions
for example). The disaggregation of seizures data is
accurate if they come from the same primary source.
Diverted (stolen or lost) firearms. Data on diverted
(stolen or lost) firearms are directly linked to the illicit
circulation of firearms and are therefore supported by
some scholars** as a possible primary indicator to
understand firearm trafficking. The challenge with
this indicator is that it does not cover all trafficked
firearms and does not measure the share of firearms
illicitly crossing borders. Data on diverted firearms
may also not have the same level of reporting require-
ments as seizures.
* For legal definitions of “seizure” and “confiscation”, see Article 2 of the
United Nations Convention against Transnational Organized Crime.
** Bromley M., Caparini M. and Malaret M., Measuring Illicit Arms and Finan-
cial Flows: Improving the Assessment of Sustainable Development Goal 16.
SIPRI Background Paper, July 2019.
21
The Primary Evidence Base: Overview Of Seizures I
One factor affecting comparability relates to coverage.
Five countries reported that their data set did not cover
the entire national territory (as opposed to 57 countries
which confirmed that the entire geographical territory was
covered), while 17 countries reported that the data did
not cover the operations of all authorities in charge of seiz-
ing firearms (as opposed to 43 countries which confirmed
that all relevant institutions were covered).
1
Incomplete
coverage was also related to the nature of the seizure cases.
Some countries, for example, submitted only the seizures
made by the national authorities in charge of tracing.
Another important element in analysing seizures across
countries is the distinction between administrative and
criminal seizures. While the aim of the UNODC data
collection was to focus primarily on seizures made in a
criminal context, countries could not always make this
distinction or clearly characterize the nature of the seizures
that they reported to UNODC. Sixteen countries clearly
distinguished between criminal and administrative seizures
and provided separate statistics.
2
In some other cases,
countries informed that administrative seizures were not
included, as seizures based on purely administrative
grounds were reportedly not possible in the national con-
text; however, there were also 8 countries for which the
data did include administrative seizures, but they could
not be distinguished from criminal seizures.
Thirdly, some countries reported figures that may have
included firearms which were recovered in ways other than
seizures, for example found and surrendered firearms.
For these reasons, the absolute value of the total number
of seized weapons, and especially comparisons among
countries on this basis, need to be interpreted with
caution.
Figures on national seizure aggregates can be contextual-
ized by taking into account the population of the country.
This approach does not address the comparability issues
1 An additional level of uncertainty was due to the fact that not all
countries which provided the data also provided information on the
extent of geographical or institutional coverage.
2 The requested data on administrative seizures was limited to the total
number of arms seized.
TABLE 1
Availability of seizure data,* by region,
2016-17 (number of countries)
* Total number of arms seized.
two parallel aspects: the extent of the illicit phenomenon
as well as the extent of a certain, targeted form of response
on the part of the authorities.
Moreover, seizures can be made on various grounds, not
all of which are necessarily linked to illicit trafficking. For
example, depending on the national context, firearms may
be seized for minor offences deemed in some countries to
be of an administrative nature - usually violations of regu-
lations pertaining to aspects such as the renewal of licences,
proper storage, restrictions on the carrying of firearms,
etc. Firearms may also be seized in the context of the com-
mission of other crimes such as homicide and robbery,
which, although serious, may not necessarily be commit-
ted with illicitly sourced firearms. Moreover, certain seri-
ous offences may also be related to the illicit nature of
firearms but still not directly to trafficking, such as the
unauthorized manufacture of firearms in a workshop, or
the removal or alteration of markings whose purpose is to
enable identification, accountable weapons management
and tracing.
For these reasons, the use of seizure data to understand
the nature of firearms trafficking needs careful considera-
tion. Seizure data alone can be difficult to interpret and
can potentially lead to misleading conclusions when taken
in isolation, because they reflect priority and capacity of
law enforcement as much as supply. This challenge can be
overcome by taking into account other, independent
sources of information. Another aspect to consider when
analysing seizures is the quality of the data which can
relate, for example, to the extent of coverage of data
reported by a given country. In cases of incomplete cover-
age, the comparability of total aggregates across countries
is limited, but this challenge can be partially overcome by
expressing the corresponding figures in relative terms
(shares of a total). When historical data are available, a
comparison of trends across countries may still be mean-
ingful even if the absolute values present issues of
comparability.
Seizure data, including different disaggregations of seized
items as well as the criminal context of seizures, are the
primary evidence base for this study. This section begins
by giving an overview of the seizure data available to
UNODC.
Through the Illicit Arms Flow Questionnaire, countries
were requested to provide data on the total number of
seized arms, their parts and components, and ammuni-
tion. Overall UNODC was able to compile data on the
number of arms seized in 2016 or 2017 for a total of 81
countries. The total number of such arms amounted to
around 550,000 in each of 2016 and 2017. However, this
figure varied greatly from country to country and the com-
parability across countries is subject to a number of
factors.
2016 2017 2016/2017
Africa 16 17 18
Americas 26 24 26
Asia 9 10 10
Europe 26 26 26
Oceania 1 1 1
Total 78 78 81
22
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 1
Total number of arms seized (logarithmic scale), by country and type of overage, 2016-17
(average)
¹ For Canada, Ghana and Paraguay data were available for 2016 only.
² For Cameroon, Guinea and Kyrgyzstan data were available for 2017 only.
Sources: UNODC IAFQ and other official sources.
FIG. 2
Total number of arms seized per 100,000 population, by country and type of coverage,
2016-17 (average)
¹ For Canada, Ghana and Paraguay data were available for 2016 only.
² For Cameroon, Guinea and Kyrgyzstan data were available for 2017 only.
Sources: UNODC IAFQ and other official sources.
related to coverage and the inclusion of administrative
seizures, but it provides a better indication of the intensity
of seizures and the possible impact they may have in single
countries. The variability in the resulting prorated figures
is significantly reduced from the variability of simple
totals, but it is still very high, with the values typically
ranging between 0.5 and 69 arms per 100,000 persons.
3
Some small countries emerge among the ones with the
3 10
th
and 90
th
percentiles, respectively.
1
10
100
1,000
10,000
100,000
1,000,000
Australia
Canada¹
ArgenƟna
Russian FederaƟon
Spain
Kenya
Mexico
Greece
Italy
Guatemala
Peru
El Salvador
Ecuador
Sudan
Slovakia
Denmark
Luxembourg
Tunisia
Japan
Jamaica
Albania
Lithuania
Central African Republic
Cote d'Ivoire
Suriname
Morocco
Burundi
Togo
Qatar
Paraguay¹
Saint Lucia
Liechtenstein
AnƟgua and Barbuda
Grenada
United States of America
Costa Rica
Portugal
Honduras
Brazil
Ukraine
CroaƟa
Lebanon
Dominican Republic
United Kingdom
Algeria
Burkina Faso
Hungary
Libya
Azerbaijan
Guyana
Cameroon
²
Romania
Republic of Moldova
Nepal
Philippines
Gh ana¹
Colombia
Angola
Belarus
Chile
Kazakhstan
Netherlands
Uruguay
Belgium
Serbia
Panama
Sweden
Norway
Montenegro
Bahamas
Slovenia
Myanmar
Cabo Verde
North Macedonia
Bolivia (PlurinaƟonal State of)
Kyrgyzstan²
Guinea²
Botswana
Tajikistan
MauriƟus
Cuba
FULL COVERAGE INCOMPLETE COVERAGE NO INFORMATION ON COVERAGE/UNCLEAR
Number of arms (logaithmic scale)
Does not include administraƟve seizures
EŽŝŶĨŽƌŵĂƟŽŶŽŶĂĚŵŝŶŝƐƚƌĂƟǀĞƐĞŝnjƵƌĞƐͬƵŶĐůĞĂƌ
Includes administraƟve seizures
0
50
100
150
200
250
300
Luxembourg
Australia
Qatar
Canada¹
Liechtenstein
Greece
ArgenƟna
El Salvador
Suriname
Guatemala
Jamaica
Spain
Slovakia
Saint Lucia
Albania
Kenya
Denmark
AnƟgua and Barbuda
Ecuador
Russian FederaƟon
Peru
Lithuania
Italy
Grenada
Tunisia
Central African Republic
Mexico
Sudan
Burundi
Togo
Cote d'Ivoire
Japan
Paraguay¹
Morocco
Costa Rica
United States of America
Portugal
Honduras
CroaƟa
Lebanon
Guyana
Dominican Republic
Ukrai
ne
Libya
Hungary
Azerbaijan
Algeria
Brazil
Burkina Faso
United Kingdom
Republic of Moldova
Cameroon²
Romania
Nepal
Ghana¹
Philippines
Uruguay
Bahamas
Belarus
Montenegro
Colombia
Angola
Cabo Verde
Chile
Belgium
Panama
Netherlands
Kazakhstan
Serbia
Slovenia
Sweden
Norway
North Macedonia
Botswana
MauriƟus
Kyrgyzstan²
Bolivia (PlurinaƟonal State of)
Guinea²
Myanmar
Tajikistan
Cuba
FULL COVERAGE INCOMPLETE COVERAGE NO INFORMATION ON COVERAGE/UNCLEAR
Number of arms seized per 100,000 persons,
2016-17 (average)
Does not include administraƟve seizures
EŽŝŶĨŽƌŵĂƟŽŶŽŶĂĚŵŝŶŝƐƚƌĂƟǀĞƐĞŝnjƵƌĞƐͬƵŶĐůĞĂƌ
Includes administraƟve seizures
23
The Primary Evidence Base: Overview Of Seizures I
MAP 1
Total number of arms seized, by country, 2017
Sources: UNODC IAFQ and other official sources.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
Dashed lines represent undetermined boundaries.
The dotted line represents approximately the Line of Control in Jammu and Kashmir agreed upon by India and Pakistan. The final status of Jammu and
Kashmir has not yet been agreed upon by the parties. The final boundary between the Republic of Sudan and the Republic of South Sudan has not yet
been determined. A dispute exists between the Governments of Argentina and the United Kingdom of Great Britain and Northern Ireland concerning
sovereignty over the Falkland Islands (Malvinas).
highest numbers of arms seized per capita. For some other
countries, the adjusted per capita metric confirms that the
extreme high or low levels are not attributable to their size
only.
In addition to the total number of arms seized, some coun-
tries provided also information on the number of cases
(instances or incidents) in which the arms were seized.
Using this information, it is possible to construct the typi-
cal number of arms seized in a given case providing addi-
tional insight into the nature of the offence. Nineteen
countries provided this kind of information, correspond-
ing to a total of 136,362 arms seized in 97,320 cases, and
overall average of 1.4 arms seized per case. Considering all
of these together, this implies that at least 60 per cent
(probably more) of these cases involved only one arm, and
at most 20 per cent (probably less) involved 3 arms or
more. Only two countries (Paraguay and Hungary) regis-
tered an average of more than 4 arms per case.
Countries were also asked to report information of signifi-
cant seizures on a case-by-case basis. The suggested criteria
for a seizure case to qualify as “significant” were any of the
following: more than 5 arms seized; involvement of organ-
ized crime groups; or the context of transnational traffick-
ing of arms. Independently of the reason for qualifying a
seizure as “significant”, the information collected on a case-
by-case basis included the types and number of arms seized
(along with other specifics such as modus operandi, other
items seized together with the firearms and information
on provenance).
Information of this kind was provided by 32 countries,
which collectively reported 357 cases. Not surprisingly,
significant seizure cases were typically much larger than
the average seizure cases in general. For example, Brazil
reported 10 significant cases averaging 26 arms per case,
compared with an average of 1.3 arms seized per case con-
sidering the total number of seized arms and total number
of cases.
Based on significant cases only, a greater variability (in
comparison with cases overall) across countries could be
observed in the typical number of arms seized per case,
with 7 countries reporting an average of 30 or more arms
seized in a handful of significant cases (3 or fewer), and at
the other extreme, 6 countries reporting less than 5 arms
seized per significant seizure (on average). Once more, this
is not surprising in view of the fact that significant seizures
constitute a small exceptional subset of a bigger
universe.
Overall, among all significant seizure cases, about a third
consisted of seizures of 5 arms or less (in addition to
ammunition and other items) and another third of seizures
of 5-10 arms, with the remaining cases ranging between
11 and more than 6,000 arms. Seizures of a single arm
were the most frequent, accounting for more than a fifth
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Total number of arms seized
100
100 - 00
01 - 000
001 - 10 000
10 000
No data a aila le
24
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
MAP 2
Total number of arms seized per 100,000 population, by country, 2017
Sources: UNODC IAFQ and other official sources.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
Dashed lines represent undetermined boundaries.
The dotted line represents approximately the Line of Control in Jammu and Kashmir agreed upon by India and Pakistan. The final status of Jammu and
Kashmir has not yet been agreed upon by the parties. The final boundary between the Republic of Sudan and the Republic of South Sudan has not yet
been determined. A dispute exists between the Governments of Argentina and the United Kingdom of Great Britain and Northern Ireland concerning
sovereignty over the Falkland Islands (Malvinas).
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Ç
Arms seized per 100,000 population
≤ 5.0
5.1 - 25.0
25.1 - 50.0
50.1 - 75.0
> 75.0
No data available
The boundaries and names shown and the designaons used on this map do not imply ocial endorsement or acceptance by the United Naons. Dashed lines represent undetermined boundaries. Doed line represents approximately the Line of Control in Jammu and Kashmir
agreed upon by India and Pakistan. The nal status of Jammu and Kashmir has not yet been agreed upon by the pares. The nal boundary betwe
en the Sudan and South Sudan has not yet been determined.A dispute exists between the Governments of Argenna and the United
Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
FIG. 3
Average number of arms seized per seizure case, based on total seizures and based on
significant seizures by country, 2016-17
* In addition, Croatia reported a third case involving only explosives.
Note: Number in brackets indicates the number of seizure cases.
Source: UNODC IAFQ.
2.4
2.9
3.4
12.3
26.5
95
110
362
6,435
1
10
100
1,000
10,000
100,000
Honduras (8517)
Republic of Moldova (61)
Mauritius (34)
Australia (44095)
Brazil (5635)
Serbia (2485)
United Kingdom (1417)
Spain (14082)
Netherlands (7775)
Albania (664)
North Macedonia (130)
Philippines (22)
Myanmar (246)
Mexico (7150)
Slovakia (938)
Norway (302)
Portugal (3744)
Paraguay (3)
Hungary (20)
Grenada (16)
Nepal (27)
Cuba (6)
Japan (8)
Philippines (29)
Portugal (3)
Ghana (3)
Albania (8)
Azerbaijan (1)
Panama (3)
Slovakia (2)
Croatia* (2)
Slovenia (3)
Lithuania (2)
Guatemala (8)
Algeria (7)
Costa Rica (1)
Honduras (50)
Paraguay (3)
Australia (71)
Sudan (45)
Denmark (5)
Myanmar (3)
Spain (28)
Brazil (10)
Bolivia (Plurinational State of) (2)
North Macedonia (3)
Mexico (3)
Romania (1)
El Salvador (1)
Tunisia (1)
Greece (1)
AGGREGATE DATA SIGNIFICANT SEIZURES REPORTED ON CASE-BY-CASE BASIS
Arms per case (log scale)
Sum of arms
Arms per case (aggregate data)
Arms per case (significant cases only)
2.4
2.9
3.4
12.3
26.5
95
110
362
6,435
1
10
100
1,000
10,000
100,000
Honduras (8517)
Republic of Moldova (61)
Mauritius (34)
Australia (44095)
Brazil (5635)
Serbia (2485)
United Kingdom (1417)
Spain (14082)
Netherlands (7775)
Albania (664)
North Macedonia (130)
Philippines (22)
Myanmar (246)
Mexico (7150)
Slovakia (938)
Norway (302)
Portugal (3744)
Paraguay (3)
Hungary (20)
Grenada (16)
Nepal (27)
Cuba (6)
Japan (8)
Philippines (29)
Portugal (3)
Ghana (3)
Albania (8)
Azerbaijan (1)
Panama (3)
Slovakia (2)
Croatia* (2)
Slovenia (3)
Lithuania (2)
Guatemala (8)
Algeria (7)
Costa Rica (1)
Honduras (50)
Paraguay (3)
Australia (71)
Sudan (45)
Denmark (5)
Myanmar (3)
Spain (28)
Brazil (10)
Bolivia (Plurinational State of) (2)
North Macedonia (3)
Mexico (3)
Romania (1)
El Salvador (1)
Tunisia (1)
Greece (1)
AGGREGATE DATA SIGNIFICANT SEIZURES REPORTED ON CASE-BY-CASE BASIS
Arms per case (log scale)
Sum of arms
Arms per case (aggregate data)
Arms per case (significant cases only)
25
The Primary Evidence Base: Overview Of Seizures I
FIG. 4
Distribution of significant seizure
cases by number of arms seized,
2016-17
Source: UNODC IAFQ.
of cases (see Figure 4). It should be noted that the spike
in the number of cases of seizures of 6 or 7 arms is likely
driven by the fact that the presence of more than 5 arms
was a suggested criterion (among others in the question-
naire) for the designation of a seizure case as
“significant”.
Criteria other than the mere number of seized arms were
also taken into account by countries. For example, all the
significant cases reported by Japan were linked to organ-
ized crime groups (Boryokudan). Albania included the
FIG. 5
Average distribution* of reported seized arms, by type, 2016-17
*Simple average based on data for 81 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms
or SALWs.
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Note: Percentages on the right hand side are renormalized so that
firearms/SALWs add up to 100%
Source: UNODC IAFQ.
Other undistinguished weapons
(unspecified/aggregate)**
Pneumatic, blank-firing
and gas weapons***
Weapons or related
items other than
firearms/SALWs,
including:***
Unclassified/
unknown
Pistols,
39%
Rifles,
18%
Revolvers,
14%
Submachine
guns, 3%
Machine
guns, 1%
Other firearms
Other SALWs
Shotguns,
25%
Firearms/
SALWs,
72%
considerable value in the black market of illicit items
among the characteristics defining a “significant” case
(along with the involvement of organized crime groups
and the international trafficking of arms).
4
Other criteria
which were explicitly mentioned for considering seizures
as significant included the international dimension and
the nature of the arms involved.
Types of firearms seized
For most countries where any seizure data were available,
a breakdown of the seizures by type was also available. The
distribution by type yields insights into the nature of the
illicit firearms market as well as firearms used for criminal
purposes in the corresponding countries. Combined with
additional complementary data such as crime statistics or
homicide data, the information on firearms seizures by
type of firearms can provide meaningful insights on the
criminological context of the seizures and represent a valu-
able investigative lead for law enforcement authorities to
better gear and prioritize their actions.
Even if reported seizures capture only a small subset of
seized arms, they represent a cross-section of all seized
arms. Hence, breakdowns of reported seizures by type of
arm (expressed as percentages for each type) may still cap-
ture the distribution of arms seized overall in the given
country to the extent the subset is representative of the
whole. Moreover, such distributions are not subject to the
variability in the magnitude of seized quantities across
countries, which is influenced by many factors, including
comprehensiveness of coverage but also the size of report-
ing country. Thus such breakdowns provide meaningful
insight into the universe of seized arms and also lend them-
4 Some of the items seized by Albania, alongside firearms, included large
quantities of cannabis, vehicles and cash.
0%
5%
10%
15%
20%
25%
0
10
20
30
40
50
60
70
80
90
1 10 100 1,000
WĞƌĐĞŶƚĂŐĞŽĨĂůůƐŝŐŶŝĮĐĂŶƚĐĂƐĞƐ
EƵŵďĞƌŽĨƐŝŐŶŝĮĐĂŶƚĐĂƐĞƐ
Size of case
(number of arms seized in case, logarithmic scale)
1 arm,
88 cases
2 arms,
17 cases
6 arms,
43 cases
7 arms,
27 cases
362 arms,
1 case
6,435 arms,
1 case
26
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
selves better to aggregations at regional and global level,
in that they are less prone to the comparability issues
inherent in the absolute values.
In single countries, pistols tend to be, on average, the most
widely seized type of firearm. This pattern is most pro-
nounced in the Americas, where handguns generally (pis-
tols and revolvers) are conspicuous (see Figure 6). Shotguns
emerge most prominently in Africa and in Asia, while rifles
consistently made up a non-negligible share of seizures in
all regions. The types of arms seized in Europe were rela-
tively diversified, including a significant proportion of
weapons other than firearms or SALWs, in particular pneu-
matic, blank-firing and gas weapons.
5
This may be indica-
tive of the relative importance in this region of conversion
of such weapons into illicit firearms.
6
Some of these regional patterns are brought into focus by
an examination of the highest proportions of specific types
of firearm reported as seized at country level (based on
data for 2016-17). This confirms the prominence of hand-
guns in the Americas, with some Caribbean countries reg-
istering unusually high proportions of pistols, and other
5 See Regional Annex.
6 While the main focus of the study is on firearms/ SALWs, some
countries also reported on other categories of arms, such as pneumatic,
blank-firing and gas weapons. It largely depends on national legisla-
tion whether these items are considered as firearms or not. However,
for the purpose of the study, only weapons falling under the definition
of the UN Firearms Protocol and the 1997 UN Panel of Governmen-
tal Expert on Small Arms were considered as firearms/SALWs. Given
that there are documented cases where some of these other weapons
were subject to conversion and turned into firearm, the study none-
theless reflects these types of arms and analyses the received data.
FIG. 6
Average distributions* of seized fire-
arms/SALWs by type, according to
region, 2016-17
* Simple averages, adjusted for any firearms/SALWs which could not be
classified and quantified into the respective category (including weapons
whose type was unknown, not reported, or reported under “Other”
without sufficient information to allow further classification) and weap-
ons other than firearms/SALWS.
Sources: UNODC IAFQ and other official sources.
countries in Latin America registering high proportions
of revolvers. Some countries in Southern Europe (Albania,
Croatia, Serbia, Slovenia) seized high proportions of rifles,
while some Eastern European countries (Hungary,
Ukraine), Northern European countries (Denmark,
Norway, Sweden), Croatia and the Netherlands stood out
in terms of the proportions of machine or submachine
guns.
7
Shotguns were most conspicuous in African countries
(Algeria, Burkina Faso, Togo, Tunisia); in addition, a high
proportion of shotguns was registered in Greece, driven
by a single seizure of 6,404 shotguns intended for the
Democratic Republic of the Congo. Moreover, high pro-
portions of high-powered arms could also be observed,
albeit in a sporadic fashion, in some African countries,
such as rifles in Kenya and Libya and machine guns in
Tunisia.
8
When considering instances of a pronounced presence of
a specific type of firearm among seizures in a given coun-
try, it is also useful to bear in mind the different kinds of
7 These proportions were notable in comparison with other coun-
tries; however in each of the mentioned countries the proportion of
machine guns or submachine guns (separately) did not exceed 16 per
cent of the total arms seized in that country over 2016-17.
8 See Regional Annex for illustrations of the national distributions of
seized arms by type.
FIG. 7
Average number of firearms of a given
type typically seized* by customs in a
single case** (excluding cases with
atypical very large numbers of fire-
arms), according to specific type of
firearm, 2016-17
* Given that some isolated seizures of large numbers of firearms were
recorded, extreme values were excluded in the computation of the aver-
ages. These values are more representative of seizures usually made on
a regular basis. The extreme values were determined on the basis of
cases between the 2nd and 98th percentile (trimmed means).
** Only cases in which at least one firearm of the respective type were
seized are considered.
Source: World Customs Organization.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Revolvers
Pistols
Shotguns
Rifles
Submachine
guns
Machine
guns
Number of firearms per case
28%
52%
34%
35%
7%
22%
8%
16%
27%
71%
24%
10%
11%
38%
14%
37%
22%
20%
8%
0% 20% 40% 60% 80% 100%
Africa
(18 countries)
Americas
(26 countries)
Asia
(10 countries)
Europe
(26 countries)
Oceania
(1 country)
Pistols Rifles Revolvers
Shotguns Submachine guns Machine guns
27
The Primary Evidence Base: Overview Of Seizures I
activity that are usually associated with specific types of
firearms. For example, shotguns are ideal for use against
small, fast-moving targets, which renders them useful for
hunting—although they are also used for law enforcement
and military purposes. Hunting rifles also exist, while
other kinds of rifles, as well as machine guns, are designed
for military use. Handguns (pistols and revolvers) are easy
to conceal and to manoeuvre at close quarters; hence they
are frequently used in the commission of violent crime.
Submachine guns are typically used by military or law
enforcement agencies.
There appear to be differences across the types of seized
firearms also in relation to the quantities in which they
are typically seized. Such patterns can be observed from
seizure cases made by customs authorities. Similarly to
seizures in general (within or across borders), the great
majority of seizure cases at the border involve one single
arm, with a few cases involving large quantities of arms.
However, based on all customs seizures recorded in the
World Customs Organization’s Customs Enforcement
Network (CEN) database, and setting aside some isolated,
very large seizures (notably of rifles), a pattern emerges
which suggests that, despite the fact that the more high-
powered weapons such as machine guns and submachine
FIG. 8
Firearms seized in typicalª customs
seizure cases, by size
b
of seizure case,
(excluding cases with atypical very
large numbers of firearms), 2016-17
a
Size is measured as the number of firearms of the given specific type
seized in the corresponding case.
b
Isolated cases of a very large size were excluded; these were defined
as cases with a size above the 98th percentile for the corresponding
specific type of firearm.
Source: World Customs Organization.
guns are seized in smaller numbers overall, in instances
when they are seized they typically come in larger quanti-
ties (see Figure 7).
More detailed breakdowns confirm these differences across
types of firearms and the general progression from hand-
guns on the one hand to machine guns and submachine
guns on the other. In terms of the number of instances,
for example, only 11 per cent of seizure cases of revolv-
ers—on one extreme—involved more than one such fire-
arm at a time, compared with 21 per cent in the case of
shotguns and—at the other extreme—almost one half in
the case of machine guns and submachine guns. This pat-
tern is amplified in terms of the number of firearms seized
(see Figure 8).
Assuming that seizures reflect the pattern of underlying
trafficking, these data suggest that while machine guns
and submachine guns may be involved in a small percent-
age of trafficking cases, they are more likely to be subject
to cross-border, well organized and larger trafficking opera-
tions. At the same time, handguns may be more likely to
be carried for personal use (licit or illicit).
9
Some cross-border cases of seizures of rifles were excep-
tionally large, indicating that the transnational trafficking
of rifles may involve few cases but of very large quantities,
much larger than other types of arms. Given that rifles
may be used in conflict situations, the exceptionally large
seizures may indicate rifle trafficking linked to conflict
areas.
Parts and components
and ammunition
Parts and components of firearms are also subject to inter-
national control; not only can they be used to replace ele-
ments of firearms, but also to (illegally) modify weapons
and even to assemble firearms in their entirety. Parts and
components can also be trafficked and hence seized.
The United Nations Firearms Protocol defines parts and
components as “any element or replacement element spe-
cifically designed for a firearm and essential to its opera-
tion, including a barrel, frame or receiver, slide or cylinder,
bolt or breech block, and any device designed or adapted
to diminish the sound caused by firing a firearm”.
Parts and components, being smaller than firearms, may
be easier to traffic and conceal. Moreover, there are fre-
quently discrepancies across countries in terms of the leg-
islation regulating parts and components; in particular,
9 In many countries, machine guns, submachine guns and certain types
of rifles are not available in the legal market for civilian use. Hence,
seizures of such arms (whether domestic or of a cross-border nature)
are in general less likely to be seized from their legitimate owner and
more likely to be linked to criminal offences, including illicit posses-
sion and illicit firearms trafficking – independently of the number of
arms seized.
1 revolver
1 pistol
1 shotgun
1 rifle
1 machine gun/
submachine gun
2
2
2
2
2
3
3
3
3
3
4-5
4-5
4-5
4-5
6+
6+
6+
6+
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Revolvers
Pistols
Shotguns
Rifles
Machine guns
and submachine
guns
1 firearm 2 firearms
3 firearms 4-5 firearms
6 firearms or more
28
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 10
Association between seizures of parts
and components and condition of
seized arms, 2017
Source: UNODC IAFQ.
Philippines
Brazil
Myanmar
Nepal
Greece
Azerbaijan
Lithuania
Australia
Angola
Peru
Sweden
United
Kingdom
Spain
Argentina
Tunisia
Grenada
Guatemala
Kenya
Romania
0
5
10
15
20
0 5 10 15 20
Proportion of industrially manufactured arms
among seizures, 2016-17, ranked
(1=lowest, 19 = highest)
Number of parts and components seized
per arm seized, 2017, ranked
(1=lowest, 19=highest)
some parts and components may be subject to restrictions
in one country but not another. This leads to the potential
for “grey trafficking” whereby a legal purchase in one coun-
try can be used to illegally supply parts and components
in another, with a reduced risk of drawing the attention
of law enforcement, including through the use of parcel
deliveries and international online purchases.
Data on total seizures of parts and components for 2016
or 2017 (or both years) were available for 33 countries,
amounting to approximately 19,000 parts and compo-
nents in 2016 and 15,000 parts and components in 2017.
In order to place these figures in perspective, it is useful
to relate them to the number of arms seized in the corre-
sponding country. For most countries which provided
these data, far fewer parts and components were seized
than entire arms - typically around 5 per cent of the
number of arms seized.
The data on parts and components corroborate, to a cer-
tain degree, an overall link between the prevalence of parts
and components and their use in assembling, adapting or
otherwise manufacturing firearms. Data from countries
that reported on both parts and components and on the
condition of seized arms, show that some of the highest
levels of seizures of parts and components (relative to the
numbers of arms seized) can go along with significant
levels of illicit manufacture—including illicit “craft” man-
ufacture (e.g. in Nepal
10
) and assembly—displayed by
10 Small Arms Survey, The Highway Routes - Small Arms Smuggling in
Eastern Nepal, Issue Brief No. 4, November 2014.
lower proportions of arms seized in “factory condition”.
In other words, it appears that in countries with a relatively
high incidence of illicit manufacture or adaptation, parts
and components are more frequently seized.
The United Nations Firearms Protocol defines “ammuni-
tion” as “the complete round or its components, including
cartridge cases, primers, propellant powder, bullets or pro-
jectiles, that are used in a firearm, provided that those
FIG. 9
Seizures of parts and components in relation to seizures of arms, by country, 2017
* 25th -75th percentile (interquartile range).
Source: UNODC IAFQ.
0.05
0.001
1.000
Kenya
Costa Rica
Grenada
Guatemala
Romania
Japan
North Macedonia
United States of America
Tunisia
Argentina
Spain
United Kingdom
Sweden
Qatar
Peru
Angola
Morocco
Croatia
Algeria
Australia
Lithuania
Libya
Azerbaijan
Greece
Nepal
Myanmar
Ecuador
Brazil
Slovenia
Philippines
Guyana
Ratio of seized parts and components
to seized arms (logarithmic scale)
Common range*
Typical value (median)
Ratio < 1 Fewer parts and components than arms seized
Ratio > 1 More parts and components than arms seized
29
The Primary Evidence Base: Overview Of Seizures I
FIG. 11
Seizures of ammunition in relation to seizures of arms, by country, 2017
* 25th -75th percentile (interquartile range).
Note: Countries with less than 5 arms seized in 2017 are excluded.
Source: UNODC IAFQ.
FIG. 12
Relative frequency of seizures of weapons and related items by customs, by type, 2016-17
Source: World Customs Organization.
23
0.01
1.00
100.00
10,000.00
Slovakia
Argentina
Croatia
Honduras
Libya
Spain
Nepal
Kenya
Sweden
Philippines
Grenada
Angola
Guatemala
Saint Lucia
Portugal
Guyana
Ecuador
Belarus
Hungary
Panama
Azerbaijan
Lithuania
Japan
Morocco
Peru
Antigua and Barbuda
Republic of Moldova
Cote d'Ivoire
North Macedonia
Greece
Ukraine
Myanmar
Brazil
Albania
Slovenia
Qatar
Central African Republic
Mexico
Sudan
Tunisia
Australia
Algeria
Bolivia (Plurinational State of)
Romania
Ratio of seized rounds of ammunition to seized arms
(logarithmic scale)
Common range*
Typical value (median)
Ratio > 1
More rounds of
ammunition than
arms seized
Ratio < 1
Fewer rounds of
ammunition than
arms seized
54%
43%
14%
10%
4%
3%
1%
0%
18%
0%
0%
10%
20%
30%
40%
50%
60%
Ammunition and
parts thereof
Any firearms/
SALWs
Pistol
Rifle
Revolver
Shotgun
Machine gun
Sub-machine gun
Other firearms
Other SALWs
Parts and components
Pneumatic, blank-firing
and gas weapons
Explosives
and propellants
Accessories
Explosive devices and
parts thereof
Explosives
and propellants
Other weapons
Other
Proportion of cases among all those
involving weapons or related items
components are themselves subject to authorization in the
respective State Party”.
The international control measures on ammunition are
generally weaker than the ones applicable to firearms.
However, while firearms are durable goods, ammunition
is expendable and hence, in a sense, more prone to traf-
ficking, as its ongoing use immediately creates a need for
it to be replaced.
Data on seizures of ammunition in 2016 or 2017 (or both
years) were available for 45 countries, amounting to a total
of 3.0 million rounds in 2016 and 7.9 million rounds in
2017. For a single country, the number of rounds of
ammunition recovered in 2017 ranged from less than 10
to more than 5 million, but typically stood at around 23
rounds of ammunition per arm seized (see Figure 11).
30
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Data from the World Customs Organization enable a com-
parison of the extent of illicit movements of ammunition
as opposed to firearms. Among all Customs seizure cases
made in 2016-17 involving weapons or related items, and
recorded in the Customs Enforcement Network database,
54 per cent involved ammunition or parts thereof, while
43 per cent involved firearms/SALWs and 14 per cent
involved parts and components. Some seizures involved
more than one element.
31
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
CHAPTER 2
FROM SEIZURES
TO TRAFFICKING
Chapter overview
This chapter delves deeper into seizure-related data, ana-
lysing the legal justifications employed for firearms seizures
and the broader criminal context in which seizures take
place. Data on the legal justification used by law enforce-
ment authorities to seize firearms and the subsequent crim-
inal context that emerges from the investigation suggest
that the great majority of firearms are first stopped on the
ground of illegal possession but trafficking is frequently
recorded at a later stage, as the criminal context in which
seizures took place. In fact, only around one half of arms
known or suspected to have been seized in a trafficking
context were typically seized with this legal justification.
When a firearm is seized, the authority that carries out the
seizure – usually a law enforcement agency – is required
to provide a legal justification for their action. Looking at
national averages, the most frequently used justification
is illicit possession of a firearm, accounting for nearly two-
thirds of seizures, while illicit firearms trafficking is the
stated reason in some 9 per cent of cases. It is probable,
however, that the legal justifications mainly reflect con-
siderations other than the full context of the seizure. Illicit
possession can be used to take the firearm out of circula-
tion quickly and efficiently while the different crimes asso-
ciated with the seized firearms (“criminal context”) may
come to light later. The majority of the crimes that emerge
in the criminal context still relate to firearms (mainly illicit
possession and trafficking) but there are also other crimes
that occur in relation to seized firearms such as violent
crime and drug trafficking.
Establishing the criminal context of a seizure through sta-
tistics is challenging. Data provided to UNODC are based
on suspected offences in aggregated form. These data indi-
cate that firearms trafficking is more prevalent as the con-
text of crime than what the legal justification used for the
seizures alone would imply, and that the share of arms
seized in the context of trafficking was, on average, 19 per
cent. Moreover, the data also highlight the regional differ-
ences with regard to the other forms of crime committed
in the context of arms not associated with offences related
to firearm regulations. Among these offences, for example,
violent crime was most prominent in Africa and Latin
America and the Caribbean, while in Europe, drug traf-
ficking was assessed more frequently than violent crime.
Conflict is also related to firearms trafficking because situ-
ations of armed conflict may weaken the rule of law and
impede the authorities’ ability to manage firearm stockpiles
and enforce regulations, therefore enhancing the illicit
supply of firearms.
32
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Firearms can enter the illicit market in several ways. As
most trafficked firearms are produced legally, the illicit
chain starts when firearms undergo transfers or movements
in violation of national or international laws and regula-
tions. This is called the ’point of diversion’ and can happen
in numerous ways. Firearms may be lost by or stolen from
the legitimate owners, who may have been, for example,
individuals, law enforcement or defence forces, private
security companies or international peacekeeping forces.
Legitimate owners may also have sold their firearm illicitly,
there may be unauthorized artisanal firearms manufactur-
ing in some countries, firearms could have been illegally
modified, or their unique markings could have been
altered, rendering the firearms illicit.
The chapter looks at the unique markings of firearms, as
mandated by international regulations. These markings
are crucial for enabling identification and tracing of fire-
arms, and in spite of well-known efforts by traffickers to
tamper with the markings, the vast majority of seized fire-
arms – some 85 per cent - are indeed appropriately marked.
Moreover, out of these seized and marked firearms, nearly
90 per cent were found to have been industrially manu-
factured, indicating that illicit firearms manufacture –
although acutely present in some locations – remains a
relatively limited phenomenon.
The chapter also looks at the various ways in which fire-
arms enter the illicit markets, focusing on diversion from
domestic sources in addition to trafficking from abroad.
While it is difficult to determine whether diversion took
place in the country of seizure or another country, some
evidence suggests that a significant share of seized illicit
firearms are diverted in the same country where they were
seized. This seems to hold true also for countries with
restrictive laws related to the range of available firearms
and the ease of obtaining them. The levels of legally regu-
lated firearms in a given country may also influence how
the black market for firearms is supplied.
The chapter closes with an analysis of the role of licit mar-
kets in firearms trafficking, homing in on prominent coun-
tries for firearms manufacturing. Illicit firearms flows are
complex and do not necessarily follow the licit flows. This
means that the country of origin of licit firearms – the
manufacturing country - and the country where the diver-
sion (where the illicit origin starts) and seizures take place
often do not overlap. This is clearly seen by the diversity
of regions that are identified as manufacturing and as point
of illicit origin. Europe is reported as the main region of
manufacturing of seizures made across the world; the
Americas is the continent most reported as illicit origin.
Firearms are durable goods and their circulation before
and after diversion to the black market can and often do
involve several transfers.
Why is it seized: legal
justification and criminal
context
In order to understand the context and the nature of fire-
arms seizures, it is helpful to look at the legal justification
used to justify the seizures as well as the criminal offences
associated with the seizures. The legal justification may
provide partial information on the criminal context of the
seizures, but this may not capture the entire picture,
because authorities may use justifications which are easier
to prove at the time of seizing the firearms. For example,
it is relatively easy to establish the offence of illicit posses-
sion—that is, that a firearm is held by somebody who is
not its legally registered owner at the time of seizure—or,
that the firearm was used to commit another criminal
offence (for example, a homicide). Illicit firearms may
come to light in an incidental fashion in the context of a
wide variety of criminal offences; in such cases, the offence
of trafficking is often considered a secondary (predicate)
offence and is particularly difficult to prove when it
occurred prior to the principal offence.
In such a context, offences such as illicit possession are
quick to establish and can justify the immediate seizure
while obviating the need for a parallel investigation into
the specific offence of trafficking. The information on the
illicit origin of the firearm may initially only be based on
suspicions and the relevant evidence may emerge, if at all,
only at a later stage of the process. Hence legal justifica-
tions for seizures and suspected offences related to seizures
describe two related but different aspects of the context of
seizures.
Data provided by 53 countries show these different
aspects.
1
Possession is typically the most prevalent justifi-
cation used by national authorities to seize firearms, but
in terms of suspected offences a different scenario emerges.
This information suggests that while firearms may be
seized on given legal grounds, the suspected offences point
to an overlap between these firearms and other, additional
forms of criminality. Due to the fact that the data on
criminal context are based on suspected offences, the
extent of this link is however difficult to assess in quanti-
tative terms.
Legal justification
Forty-eight countries provided information about the legal
grounds for seizures of arms. On average, countries tend
to seize the majority of arms for offences related to illicit
possession (64 per cent), followed by trafficking and illicit
use. Other specific firearm-related offences, namely altered
1 Forty-eight countries provided data on legal justification and thirty-six
countries provided data on other suspected offences. Some countries
are included among both the 48 and the 36 because they reported on
both types of data.
33
Why is it seized: legal justification and criminal context 2
firearms trafficking requires specific examinations, includ-
ing the tracing of the firearm (see Section: SDG indicator
16.4.2 and tracing) and additional investigations which
take more time and may involve other authorities. The
charge of illicit possession may be the easiest and fastest
way to take firearms out of circulation, even if it may
appear clear from additional circumstances, that the
firearm(s) had been trafficked.
The predominance of illicit possession as legal grounds
for seizure may be influenced in part by practical consid-
erations, such as the relative ease of proving such an
offence for law enforcement agencies at the time of issu-
ance of the seizure order, as well as issues of perceived pri-
ority and resource allocation.
Another complementary explanation relates to the size of
seizures. Seizures of one or a few firearms may more likely
be related to illegal possession than to trafficking. So the
fact that the largest portion of firearms are seized for illicit
possession may be linked to the prominence of small
seizures.
There is, however, some regional variation. The majority
of countries in Europe reported illegal possession as the
predominant justification for seizures, but this might relate
to the large number of cases with single arms seized. But
there are noticeable departures from this general pattern
showing the possible different nature of seizures in single
countries. A large share of firearms seized on the grounds
of trafficking was reported by Hungary - 127 arms out of
a total of 337 - and Greece, with a very large single seizure
markings and illicit manufacture, are typically much less
frequent in national seizures (on average), as measured in
terms of number of arms seized. This can be the results of
different dynamics.
As stated earlier, firearms seizures do not point automati-
cally to illicit trafficking, unless the firearms were seized
at the border or in any other context that establishes a clear
link to a trafficking case. Illegal possession may be easier
to prove than trafficking and is the offence most used by
law enforcement to justify seizures. Proving the offence of
FIG. 1
Average distribution* in a single
country of seized arms, by legal
justification, 2016-17
* Simple average of data 48 countries, adjusted for seizures whose legal
justification was unknown or unclassified.
Source: UNODC IAFQ.
Illicit
possession
64%
Arms
trafficking
9%
Illicit use
8%
Illicit
manufacture
1%
Altered markings
1%
Other
17%
International and national definitions of firearms trafficking
The United Nations Firearms Protocol, in its Article 3
(e), defines “illicit trafficking” as “the import, export,
acquisition, sale, delivery, movement or transfer of
firearms, their parts and components, and ammuni-
tion from or across the territory of one State Party to
that of another State Party if any one of the States
Parties concerned does not authorize it in accordance
with the terms of this Protocol or if the firearms are
not marked in accordance with article 8 of this Proto-
col”. This definition contains two crucial elements: (1)
a transnational character, and (2) a violation of a
regulatory measure (i.e. lacking authorization or
having improper markings). The related offences
defined by the Protocol are intended to increase
transparency associated with the cross-border move-
ment of firearms and related items.
a
National definitions of firearms trafficking of State
parties and non-State parties to the Firearms Protocol
may however vary, both in terms of nomenclature
and in terms of the punishable behaviour. Some
national legislation may include in one single offence
the illicit nature of both domestic and transnational
transfers. Countries with such legislation may not be
able to distinguish seizures related to transnational
arms trafficking from those related to domestic illicit
circulation.
Based on the information provided by Member States,
it seems that this type of legislation is quite common.
Only 6 of the 53 countries that provided information
regarding their national definitions of firearms traf-
ficking do not include the transfer or movement of
arms within national borders in their firearms traffick-
ing legislation.
It should also be noted that not all countries have
established firearms trafficking as a criminal offence,
and the relevant violation may have the character of
a customs infringement, such as “illegal importa-
tion”, “smuggling”, or “contraband”.
a See Legislative Guide for the implementation of the UN Firearms
Protocol, paragraph 201.
34
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Criminal context
Illicitly sourced or trafficked firearms may be seized in the
context of criminal activities which are not related to fire-
arms, such as drug trafficking, homicides or other crimes.
Among the 36 countries that provided information on
arms seized in the context of offences not related to fire-
arm regulations, a relatively small proportion of firearms
were seized in relation to drugs (from 0 per cent to 36 per
cent) and violent crime (from 0 per cent to 21 per cent,
with only one country reaching 79 per cent), with a neg-
ligible percentage related to other forms of organized
crime, and terrorism. Still the highest percentage of fire-
arms are seized in connection to suspected firearms-related
offences (possession and trafficking). These data confirm
that there is a connection between illegal activities related
to firearms and other forms of crime, but they do not
define the magnitude of this connection because of pos-
sible underreporting of suspicions based on observed cir-
cumstances and the fact that the starting point of the
analysis is the firearms seizures. More connections could
potentially be found if the starting point were drug traf-
ficking or other crimes.
While the average percentage of arms seized related to
forms of crime not connected to arms related offences is
low, there are regional variations in the different kinds of
crime (other than arms offences) and in the national shares
of seizures related to other forms of crime. As a context
for arms seizures, violent crime was on average most pro-
nounced in Latin America and the Caribbean and in
of trafficking of 6,435 firearms (of which 6,404 were
shotguns).
2
Sweden for example, reported a relatively high
proportion of arms (284 out of 1,769 during the period
of 2016-17) seized on the grounds of alterations in the
firearm markings.
3
Illicit manufacture as legal grounds for seizure was most
marked – even if still very small in relative terms - in
Africa, driven mainly by Algeria, which reported 265 arms
seized in 2016-17 on these grounds (18 per cent of the
total arms seized in these two years) and Kenya (424 arms
out of 5,264 seized in 2016). This is in keeping with the
known prevalence of artisanal manufacture of firearms in
this region. Burkina Faso also seized significant numbers
of artisanally manufactured arms, but data on legal justi-
fication were not provided.
On the other hand, trafficking as legal justification was
most marked in countries in Latin America and the Car-
ibbean. For example, in the Plurinational State of Bolivia,
69 arms were seized on these grounds in 2017, out of a
total of 152. In the same region, Jamaica registered the
largest proportion for illicit manufacture as a legal justifi-
cation of seizures: all of the 111 seized “homemade” arms
in 2016-17 (out of a total of 1,511) were seized with “illicit
manufacture” as a legal justification (see also Figure 8).
2 This seizure was made by the Hellenic Coast Guard in Crete, from
containers intended for the Democratic Republic of the Congo.
3 The purpose of the markings is to enable identification, accountable
weapons management and tracing.
FIG. 2
Average national distributions* of
seized arms by legal justification,
according to region, 2016-17
* Simple averages adjusted for seizures whose legal justification was
unknown or unclassified.
Source: UNODC IAFQ.
FIG. 3
Criminal context of seizures, meas-
ured by arms seized in context as
average proportion* of arms seized
in a single country, 2016-17
* Simple averages, based on data for 36 countries.
** Error bars reflect uncertainty due to potential overlaps in recording of
trafficking as a legal justification and as a criminal context.
Source: UNODC IAFQ.
0%
5%
10%
15%
20%
Arms offences, other
than trafficking
Arms trafficking**
Violent crime
Drug trafficking
Other forms of
organized crime
Other forms of
trafficking
Terrorism
Proportion of seized arms (percentage)
Arms-related offences
Other forms of crime
44%
67%
76%
14%
7%
8%
9%
8%
8%
31%
15%
7%
0% 20% 40% 60% 80% 100%
Latin America
and the Caribbean*
(16 countries)
Africa*
(9 countries)
Europe*
(16 countries)
Illicit possession Arms trafficking
Illicit use Altered markings
Illicit manufacture Other
35
Why is it seized: legal justification and criminal context 2
Africa. This is in line with the relatively high levels of vio-
lent deaths (conflict-related or otherwise), in particular
intentional homicide and specifically firearm-related hom
-
icides, known to affect these regions.
4
Drug trafficking was
also prominent in Latin America and the Caribbean, where
the links between this phenomenon and violence are also
well-documented.
5, 6
Guatemala, for example, reported that, out of 9,626 total
arms seized in 2016-17, more than three quarters (7,641)
were seized in the context of violent crime, 250 in the
context of drug trafficking and 500 in the context of other
forms of organized crime. In Burundi, although total sei-
zures were low (235 arms during 2016-17), around a fifth
were seized in the context of a violent crime, while the
analogous share was around 15 per cent in Algeria and
Morocco. In Tunisia, out of 1,570 arms seized in 2016-17,
11 per cent were linked to terrorism.
In Europe, notable quantities of arms were seized in the
context of drug trafficking by Portugal (668 during 2016-
17), Spain (538) and, in relative terms, Albania (146 out
of 963). Denmark and Sweden were among the countries
in Europe registering the highest proportions of seized
arms linked to violent crime (13 per cent in Denmark and
8 per cent in Sweden).
4 UNODC, Global Study on Homicide, 2019.
5 UNODC, Global Study on Homicide, 2019; UNODC, World Drug
Report, 2016.
6 UNODC, World Drug Report, 2016.
The data on criminal context suggest that more serious
arms offences, such as arms trafficking, are more prevalent
than would appear from the legal justification used as
grounds for seizure. Indeed, taking into account both
kinds of reporting linking arms to trafficking (suspected
criminal context as well as legal justification), the data
suggest that the proportion of arms seized which could be
placed in the context of trafficking was, in most countries,
around double that which the legal justification alone
would indicate (on average, 19 per cent in each country).
In other words, only around one half of arms known or
suspected to have been seized in a trafficking context were
typically seized on these grounds as a legal justification.
Africa registered the highest average proportions of arms
seized and linked to trafficking, with notable proportions
reported by Sudan, Tunisia, Algeria and Kenya. In Europe,
the highest percentages were registered by Romania, North
Macedonia and Greece, each of which reported more than
half of the arms seized in 2016-17 being linked to firearms
trafficking. However, the quantity of arms seized was small
in Romania (85), while the high proportion in Greece was
attributable to a single large seizure.
The seizure of a firearm is the first step of a criminal jus-
tice process which later includes investigation, prosecution
and possible conviction. The legal justification which
forms the basis of a seizure does not necessarily correspond
to the final charges and convictions resulting from the
seizures. Placing the legal justification of firearms traffick-
ing in relation to the actual extent of firearms which had
FIG. 4
Other forms of crime (not connected to
firearms offences) emerging in the con-
text of arms seizures, measured by arms
seized as proportion of overall seizures
in a given country (averages by region),
2016-17
* Simple averages.
Source: UNODC IAFQ.
0%
2%
4%
6%
8%
10%
12%
Africa*
(7 countries)
Europe*
(15 countries)
Latin America and
the Caribbean*
(9 countries)
Average proportion of seized arms seized
in the context of given crime type (percentage)
Violent crime
Drug trafficking
Other forms of organized crime
Other forms of trafficking
Terrorism
FIG. 5
Arms seized in the context (confirmed
or suspected) of trafficking, as propor-
tion* of overall seizures in a given
country (averages by region), 2016-17
* Based on data for 36 countries (simple averages).
Note: Error bars reflect uncertainty due to potential overlaps in record-
ing of trafficking as a legal justification and as a criminal context.
Source: UNODC IAFQ.
0%
5%
10%
15%
20%
25%
30%
35%
Latin America
and
the Caribbean
(9 countries)
Africa
(7 countries)
Europe
(15 countries)
Proportion of seized arms (percentage)
36
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 6
Proportion of arms seized on
the basis of trafficking as a legal
justification, among trafficked arms,
2016-17
Source: UNODC IAFQ.
been trafficked, gives an initial indication of the extent to
which trafficking of firearms is detected and recognized
at the earliest stage of the criminal justice process. Based
on the available data, only about one half of arms which
could be linked to trafficking were already recognized as
such in terms of legal justification at the point of seizure.
However, given the reliance on suspicions of criminal con-
text, and on such suspicions being captured in the record-
ing and reporting, the available data on criminal context
likely underestimate the de facto levels of trafficking in
the context of seizures; consequently, the share of one half
is likely optimistic and should be taken as a lower bound.
An independent way to estimate this share is to assume,
as a worst case scenario, that all seized arms had been traf-
ficked, with the exception of those seized from their legiti-
mate owners (as these arms had presumably not been
trafficked). This “pessimistic” approach is biased in the
opposite direction but, when combined with the data on
known and suspected cases of trafficking, allows to derive
the proportion as a range. This approach would indicate
that, in Africa (based on data from 4 countries), on aver-
age 15-32 per cent of arms that were trafficked were seized
as such. In other words, less than one third of trafficked
arms were indeed seized for trafficking (as a legal justifica-
tion); similarly, the proportion is estimated to range
between 26 and 63 per cent in Latin America (based on
5 countries) and 12 to 63 per cent in the case of Southern
and Eastern Europe (based on 8 countries).
0%
10%
20%
30%
40%
50%
60%
70%
Africa
(4 countries)
Latin America
(5 countries)
Southern and
Eastern Europe
(8 countries)
Proportion among seized trafficked arms (percentage)
Upper end of range
26% - 63%
12% - 62%
15% - 34%
Proportion (range) among seized trafficked arms for
which "trafficking" was also the legal basis of seizure
Lower end of range
Understanding the black
market for firearms
Supply and demand of illicitly
sourced firearms
The demand for illicitly sourced and trafficked firearms
is linked to a multitude of factors which revolve around
three main elements: crime, conflict and speculation. Fire-
arms can also have a value by themselves and be used as
monetary value in exchange for commodities from other
illicit markets such as drugs. Each country may have a dif-
ferent factor or a combination of factors that drive demand.
In certain countries, crime may be the primary driver as
criminals opt to resort to firearms which are not accessible
to them through legal means – for example, because their
profile or criminal record precludes them from possessing
the weapon of choice, or because such a weapon is not
freely available to civilians in the country.
The intended illicit use of firearms by criminals may vary,
with perpetration of violent crime perhaps the most obvi-
ous. Firearms become an instrument of violent crime in
different ways, mainly threatening and perpetrating vio-
lence that result in homicide, assaults, robberies, extortion
or kidnapping. Firearms are instrumental to organized
crime and gangs as they can convey the sense of threat and
power which allow the protection of the group, the terri-
tory and eventually their illicit markets. In some cases,
firearms help organized crime groups and gangs to display
extreme brutality as a message intended to catch the atten-
tion of the general public as well as state authorities.
Trafficked firearms can also be instrumental to violence
perpetrated for political or ideological reasons, in particu-
lar insurgency and terrorism. Non-state armed groups
which exercise de facto control of a certain territory may
rely on diverted or trafficked firearms as tools for “enforce-
ment”. Certain groups of a legitimate nature may also
resort to firearms obtained indiscriminately, such as com-
munity-based “vigilante groups”. The various forms of
armed conflict, including civil war and armed resistance,
create a demand for firearms which may be met through
trafficking.
Diversion of legally manufactured firearms to illegal mar-
kets appears to be a major contributor driving supply of
illicit firearms.
7
The licit holdings of firearms by the civil-
ian population are susceptible to being diverted into the
black market, from where they can become instruments
of illegal activity. Loopholes in legislation and weak rule
of law may facilitate diversion. Armed conflict may drive
demand but can also drive supply, in that it may bring
about a breakdown in the rule of law and undermine the
ability of authorities to manage state stockpiles and enforce
7 See subsections on “Condition of seized firearms and grey areas” and
“Evidence on diversion”.
37
Understanding the black market for firearms 2
regulations on civilian firearms which may be more easily
diverted into illicit markets. In post-conflict scenarios, the
accumulation of weapons over the course of the preceding
conflict may also result in a pool of firearms outside of the
legal control mechanisms which can increase supply.
Independently of the presence of conflict, holdings of fire-
arms by military and law enforcement agencies can con-
stitute a risk of diversion and can increase supply if the
adequate measures are not in place to ensure proper inven-
tory management, storage, transportation and disposal and
to safeguard against leakage through theft or corruption.
Modalities for firearms to enter
the black market
Firearms trafficking presents distinct features as compared
to other commodities. Some forms of trafficking, such as
the trafficking of drugs which are under the most restric-
tive regulation in the international drug conventions (such
as cocaine and heroin), have to do with the movement and
trade of commodities which are illicitly sourced; for such
commodities, the trafficking chain typically starts at the
place of cultivation or manufacture. The licit production
of these substances is typically insignificant in comparison
with the amounts illicitly manufactured.
In contrast, the vast majority of firearms, and indeed of
trafficked firearms, are manufactured legally by licensed
manufacturers.
8
Aside from certain exceptions which
account for a minority of firearms of illicit origin, the illicit
chain starts at the moment when the legal chain is diverted
into the illegal chain – that is, when transfers or movement
occur in violation of national laws or international
provisions.
In order to understand how the black market for firearms
is supplied, it thus becomes crucial to understand the point
at which firearms cross over from the licit sphere to the
illicit sphere; this is referred to as the “point of diversion.”
Once a firearm enters the illicit market, it can in principle
continue to be trafficked indefinitely, unless taken out of
circulation; hence diversion is only the beginning of the
trafficking chain.
The transition from the licit sphere into the black market
can occur at various points of the firearm’s life cycle:
domestic or international transfer and transit, storage, pos-
session and use or final disposal. There are various modali-
ties for diversion to happen.
Many firearms seized from illicit circulation are recorded
as lost, or stolen from their legitimate holder (who may be
a private individual, national law enforcement or defence
forces, or personnel of private security companies, et cetera).
8 This is confirmed, for example, by the presence of markings found on
seized firearms – see next section, “Condition of seized firearms and
grey areas”.
Firearms may be stolen from an unaware legitimate holder,
but it is also possible for the legitimate holder to sell their
firearm to a buyer in violation of laws governing transfers,
licensing and registration, as the national legislation may
require. Indeed, one technique which has been documented
for criminals to acquire firearms is that of “straw purchases”,
9
whereby an individual with a clean criminal record buys a
firearm from a licensed outlet, quite possibly without vio-
lating any laws in so doing, precisely with the intention of
selling it either on the black market (in the same country
or not) or to a pre-determined buyer who would otherwise
not be eligible to own such a firearm.
In conflict and post-conflict countries, the accumulation
of legal stockpiles of weapons may create the potential for
firearms to reach the hands of non-state armed groups,
other criminal groups or even the general population, espe-
cially if that very conflict weakens the ability of the state
infrastructure to manage those stockpiles properly. Another
scenario, of crucial importance, is that of a firearm which
is legally acquired or held in one country and which is
then transferred illegally to another country, circumvent-
ing the applicable regulations on transfers, either by simply
concealing and transporting the firearms across borders or
by the use of falsified or misleading declarations, fake and
decoy recipients or other methods. In such cases, the point
of diversion can arguably be considered to be the interna-
tional cross-border movement itself rather than a specific
country. Moreover, this modality is especially relevant in
case of source countries which have less stringent restric-
tions on the licit market than the destination countries.
In some countries, firearms may be manufactured in arti-
sanal settings, which may be an avenue for firearms to
enter the illicit market. There are also indirect ways in
which a firearm may illicitly come into existence. Some
weapons, such as gas pistols, may in some countries fall
outside of the formal definition of a firearm and therefore
not necessarily be subject to the same legal restrictions,
but may at the same time easily lend themselves to con-
version into a functioning firearm by any person with a
modicum of skill. In some instances, it may be possible to
reactivate a deactivated firearm, another modality result-
ing in a functional firearm which thereby enters the illicit
sphere. It may also be possible to circumvent laws and
regulations by buying various parts and components of
firearms separately, potentially from different countries
with different control regimes, in order to assemble com-
plete firearms. There have also been known incidents of
the use of 3-D printing to produce firearms, although the
technology appears not to be sufficiently developed to
yield reliable firearms yet.
9 See, for example: United States, Department of the Treasury, Bureau
of Alcohol, Tobacco & Firearms, Following the Gun: Enforcing Federal
Laws Against Firearms Traffickers, Department of the Treasury, Bureau
of Alcohol, Tobacco & Firearms, June 2000.
38
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Other modalities
From industrially
manufactured firearms
Holdings of
law enforcement/military
Licensed dealers
Abroad
Manufacturers, traders,
legitimate owners
Parts and components
Other weapons
Deactivated firearms
Domestic
legally held firearms
Assembly
Conversion
Corruption, theft, inadequate
stockpile management
Modification, altering markings,
illegal transfers
Straw purchases,
theft, illegal sales
Unauthorized manufacture of
rudimentary or artisanal firearms
3D printing
Scrap metal, sheet metal,
plumbing items, tools,
nail guns
Printing
MODALITIES FOR A FIREARM TO ENTER THE BLACK MARKET
Diverted exports, fake recipients,
clandestine shipments, mislabelling
Reactivation
Black market/
“Illicit” firearms
Firearms can also be modified in ways which significantly
change their functionality and features, and thereby poten-
tially circumvent legal restrictions. For example, shorten-
ing the barrel of a shotgun makes it easier to conceal and
use at close quarters, while the potential of a semi-auto-
matic firearm for inflicting damage can be significantly
increased by rendering it capable of automatic fire. Finally,
the United Nations Firearms Protocol stipulates that the
alteration or obliteration of markings meant to enable the
identification, accountable management and tracing of
the firearm is itself an offence and renders the firearm illicit
even if it continues to be held by the person who originally
acquired it through legal means.
Condition of seized firearms
and grey areas
The majority of trafficked firearms are likely to have been
legally manufactured and diverted after manufacture along
national and international trafficking routes. Most of fire-
arms seized by countries are uniquely marked so they are
traceable. Among those that are not marked, seizures data
suggest that there is regional variation on how traffickers
conceal the firearms’ provenance or source firearms outside
of marked, legal production. In Latin America and the
Caribbean, traffickers seem to use legally manufactured
firearms and alter markings to hide their origin, while in
Africa it is more likely than in other regions to find no
markings on seized firearms, likely reflecting the impor-
tance of artisanal production in this region.
Firearms can be produced or modified in different ways.
The majority of trafficked firearms are industrially manu-
factured, but seizures data show that there are some other
methods used in clandestine production varying from fire-
arms which are modified, converted, assembled, reacti-
vated or artisanally manufactured. In some countries it is
possible to note specific characteristics of concealing meth
-
ods or illegal manufacturing, for example, artisanal manu-
facture is more common in parts of Asia and Africa.
Firearms can also be converted from other weapons such
as those designed to only fire tear gas or even blanks as well
as pneumatic (“air”) weapons, which rely on air pressure
(rather than an explosion) to convey motion to the projec-
tile. Specific patterns in some countries suggest a geographi-
cal concentration of the different methods used to convert
such weapons into firearms, most notably in Europe.
“Every firearm tells a story”
The identification and analysis of firearms and their
ammunition can give an indication of their illicit nature
and clues about specific modi operandi of criminals. With
the right procedures it is possible for example to determine
if firearms were discharged and used in crime; whether
they were used to kill or injure a particular victim. Fire-
arms that are not produced in an industrial manner and
lack original marking and serial number, or those whose
markings have been erased, are likely to have been illicitly
manufactured or diverted to the illicit market at a certain
point; but without a traceable mark, investigation on the
nature of the firearms is more difficult.
39
Understanding the black market for firearms 2
The presence of markings helps to understand the history
of firearms found in illegal markets. As per article 8 of the
United Nations Firearms Protocol, States Parties which
authorize manufacture of firearms are to “require unique
marking providing the name of the manufacturer, the
country or place of manufacture and the serial number,
or maintain any alternative unique user-friendly marking
with simple geometric symbols in combination with a
numeric and/or alphanumeric code, permitting ready
identification by all States of the country of manufacture”.
10
Such markings are crucial to facilitate the identification
and tracing of firearms and define points of vulnerability
in the illegal market; however, criminals and traffickers
sometimes attempt to obliterate such markings on indus-
trially manufactured firearms or to render them unread-
able–actions which States Parties are also required to
criminalize. Moreover, in cases of illicitly manufactured
firearms, including illicit artisanal manufacture, such
markings are typically not present.
On average, 85 per cent of arms seized in a single country
in 2016-17 were uniquely marked – reflecting the fact that
the vast majority of firearms originate from industrial legal
manufacture. Among those that were not uniquely
marked, once more, regional variations could be observed
in terms of the balance between firearms with altered
markings and no markings at all. In Latin America and
the Caribbean, arms with altered markings dominated the
firearms that were not uniquely marked, driven mainly by
the numbers reported by Argentina (9,980 arms with
altered markings out of 43,321) and Brazil (896 out of
7,078). Altered markings were also significant in Europe:
10 Other requirements are applicable in the case of importation of fire-
arms and transfers from government stocks to civilian use.
Spain and Portugal were among the most prominent in
the region, seizing respectively 2,322 and 1,275 arms with
altered markings. However, in the United Kingdom, seized
arms with no markings (622) outnumbered arms with
altered markings (111). Arms with no markings at all were
conspicuous in Africa, likely reflecting the importance of
artisanal production in this region, and driven by the
shares registered in Algeria, the Central African Republic
and Ghana (although the seizures in Ghana were very
small in absolute terms). Algeria also reported instances
of artisanal firearms seized among significant cases, includ-
ing artisanal hunting weapons.
There are several types of actors who may produce or adapt
firearms outside of the legal industrial process. This
includes hobbyists, gunsmiths and tribal groups who pro-
duce firearms for cultural reasons or for the purpose of
hunting, as well as criminals and traffickers. Some research
suggests that non-state armed groups operating in conflict
and post-conflict zones rely mainly on professionally-man-
ufactured small arms, while they may resort to their own
production of light weapons and their ammunition.
11
Some rudimentary arms can be produced by putting
together commercially available parts which were not
meant to be used to construct a firearm. In other cases,
the starting point to illicitly obtain a firearm can be any
of the following: a deactivated firearm; an “ordinary” fire-
arm which is subsequently modified to increase its effi-
ciency or capacity for damage; industrially manufactured
parts and components of firearms which are illegally
assembled; or other weapons which can be easily converted
into firearms.
11 Small Arms Survey, Beyond State Control, Improvised and Craft-pro-
duced Small Arms and Light Weapons, 2018.
FIG. 7
Average national distributions of seized arms by marking status, global and by region, 2016-17
Notes: Simple averages of distributions. Adjusted for arms whose marking status was unknown or unclassified.
Source: UNODC IAFQ.
Uniquely
marked, 82%
Altered
markings,
1%
No
markings,
17%
Africa (9 countries)
Uniquely
marked,
87%
Altered
markings,
6%
No
markings,
7%
Europe (11 countries)
Uniquely
marked,
94%
Altered markings,
5%
No
markings,
1%
Latin America and the Caribbean
(12 countries)
Uniquely marked,
85%
Altered
markings,
4%
No markings,
11%
Global (38 countries)
40
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
Based on 21 countries which gave a comprehensive break-
down of seizures according to this typology, on average 88
per cent of seized firearms at the national level were indus
-
trially manufactured in factory condition (with no signs
of alteration or deactivation); the other categories (modi-
fied firearms; firearms which had been converted (from
other kinds of weapon); firearms which had been assem-
bled (from parts and components); reactivated firearms;
or arms otherwise illicitly manufactured (including in arti-
sanal settings) together accounted for an average of less
than 15 per cent. However, the exceptions yield some
insights into the modalities employed by criminals to obvi-
ate restrictions on firearms and the ways to acquire them.
An examination of the countries with unusually high pro-
portions (compared to all countries in general – see Figure
8): of seized arms that were not industrially manufactured
(in factory condition) shows that several countries in
Europe, notably Northern Europe (Denmark, Lithuania,
Norway, Sweden and the United Kingdom) have some
specificities. The United Kingdom in particular was con-
spicuous in terms of the proportions of seized arms that
were converted, reactivated and modified; this could reflect
the relatively restrictive nature of laws and regulations in
this country, which may spur individuals to resort to adap-
tations of firearms or conversions of other, more easily
FIG. 8
Countries seizing notable proportionsª of arms other than industrially manufactured arms
in factory condition, 2016-17
a
A proportion was deemed notable if it was based on a total of at least 30 arms seized and exceeded a threshold calculated as the 90th percentile of
the proportions from countries which reported a comprehensive breakdown.
b
Includes ammunition and other items.
c
May include licensed artisanal manufacture.
d
Very few countries reported reactivated arms.
Source: UNODC IAFQ.
accessible weapons. Lithuania identified the main type of
trafficked firearms as semi-automatic pistols, mostly con-
verted and/or reactivated. Sweden and Denmark reported
significant proportions of arms converted from “starter
pistols” and from “gas-/alarm pistols”, respectively.
Ukraine registered significant proportions of assembled
and converted arms. This corroborates other evidence of
these two types of illicit manufacture in this country,
including the discovery of conversion workshops catering
mainly for criminals, and the detection in recent years of
workshops specializing in the assembly of firearms from
parts and components.
12
Some countries in Africa registered high proportions of
seized arms which were neither industrially manufactured
nor produced from industrially manufactured weapons or
components through techniques such as modification,
conversion, reactivation and assembly of components. This
reflects to a large extent the importance of artisanal pro-
duction – including under licence in some countries - of
firearms in this region.
12 Buscemi, F., Duquet, N., Golovko, E. & Woods, E. ‘Illicit firearms
proliferation in the EU periphery: The case of Ukraine’, in Duquet,
N. (Ed.), Triggering terror: Illicit gun markets and firearms acquisition
of terrorist networks in Europe, Brussels: Flemish Peace Institute, 2018.
Global
thresholdª
Global
thresholdª
Global
thresholdª
Global
thresholdª
Central
African Republic
Denmark
Norway
Philippines
Philippines
Portugal
Tunisia
United Kingdom
United Kingdom
United Kingdom
Azerbaijan
Azerbaijan
Republic of
Moldova
Republic of
Moldova
Republic of
Moldova
Lithuania
Lithuania
Lithuania
Jamaica
Burkina Faso
Algeria
Slovakia
Nepal
Sweden
Ukraine
Ukraine
1%
10%
100%
Converted
arms
Assembled
arms
Reactivated
arms
Modified
arms
Arms otherwise
illicitly manufactured,
including artisanal
production
Percentage of reported seizures (logarithmic scale)
41
Understanding the black market for firearms 2
Nepal also registered a high proportion of such firearms
among its seizures, with some described as “homemade”
pistols and typically seized individually (only one per
instance). This is again in line with the documented preva-
lence of artisanal manufacture in this country.
13
Although countries in Latin America and the Caribbean
were not particularly prominent (with the exception of
Jamaica) in terms of the proportions of illicitly manufac-
tured arms among seizures, several countries reported
“armas hechizas” (rudimentary firearms) among their sei-
zures, including Costa Rica, Honduras and Peru. Based
on historical data collected in the 2015 UNODC firearms
data collection exercise, rudimentary firearms were promi-
nent among seizures in Peru in 2010 and 2011. Despite
some comparability issues, the data for 2016-17 from the
present study suggest that this phenomenon has consider-
ably decreased. Significant numbers of arms “otherwise
illicitly manufactured” were also reported by Argentina,
El Salvador and Guatemala.
Pneumatic, blank-firing and gas
weapons
There are some weapons that do not technically qualify
as “firearms” as defined by the Firearms Protocol that
could, however, be of particular relevance to firearms traf-
ficking, in that they may be easily converted into firearms
(depending on the characteristics of the specific models).
They include weapons designed to only fire tear gas or
13 Small Arms Survey, The Highway Routes - Small Arms Smuggling in
Eastern Nepal, Issue Brief No. 4, November 2014.
FIG. 9
Seizures of artisanal/rudimentary
arms in comparison with total arms
seizures in Peru, 2010-17
* Includes all types of illicit manufacture other than conversion, assembly
and reactivation. Does not include modified arms.
Note: No data were available for 2014 and 2015.
Source: UNODC IAFQ.
0
1,000
2,000
3,000
4,000
5,000
6,000
2010
2011
2012
2013
2016
2017
Number of arms
Total
Rudimentary firearms
Artisanal production*
even blanks (e.g. starting pistols used for track and field
races) as well as pneumatic (“air”) weapons, which rely on
air pressure (rather than an explosion) to convey motion
to the projectile. Since the rules and regulations governing
the sale, transfer and ownership of such weapons are less
stringent than firearms, these weapons can be legally sold
and transferred among countries and regions, and subse-
quently be converted into illicit firearms and used in
crime. Moreover, such rules and regulations depend on
national legislation, and consequently the interpretation
of seizures data needs to be done with caution since their
legal justification (if any) varies across countries.
Several countries provided information on seizures of such
weapons. Independently of whether these weapons had
been illicitly converted, or were meant to be converted,
into firearms, the fact that they were seized and reported
indicates that they were detected in circumstances which
violated the applicable national laws and regulations.
Aside from the total number of seized pneumatic, blank-
firing and gas weapons, several countries reported on
emerging modalities, related to these weapons, which may
affect the illicit manufacture or adaptation of firearms.
Denmark reported a rise, in 2016-17, in the levels of traf-
ficking of converted gas-/alarm pistols. Trafficking of
sawn-off shotguns was also occasionally detected alongside
the more prevalent pistols, while weapons assembled from
separately trafficked parts and components were rarely
encountered. These pieces of information together suggest
that there is a demand in the criminal underworld in Den-
mark for smaller weapons that are easier to conceal.
Sweden registered three cases of 3D-printed firearms in
2017, and one in 2016.
North Macedonia mentioned some cases of individuals
who held gas pistols, signal pistols or an air rifle without
adequate permission, and highlighted the trade in the ille-
gal market of flare guns illegally converted into other types
of weapons, including semi-automatic weapons. These
weapons were reportedly most frequently procured from
Turkey, where they were manufactured, but it was also
possible to procure them from Bulgaria. The flare guns
were usually converted by replacing the barrel, after which
the weapons could be used as firearms with 6.35mm and
7.65mm calibre rounds.
Hungary reported an increase of incoming gas and alarm
pistols from Turkey as well as an increase in such weapons
purchased domestically (the legislation in Hungary does
not require permission to purchase gas or alarm pistols
nor does it require such weapons to be registered). This
trend is likely driven by the demand for such weapons for
the purpose of illegal conversion.
Portugal highlighted the conversion of 6.35mm calibre
weapons into firearms as well as the modification of shot-
guns whose barrels are cut, and also reported indications
42
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 10
Reported seizures¹ of pneumatic, blank-firing and gas weapons, 2016-17
1
Includes only cases where the number of seized weapons of this type could be quantified. Several other countries reported seizures of such weapons
aggregated with other types.
2
Figures for Japan may also include machine guns and other weapons.
3
Figures for the United Kingdom potentially include found weapons in addition to seized weapons.
4
Data for the United States include firearms (as classified by the Bureau of Alcohol, Tobacco, Firearms and Explosives, United States) submitted for trac-
ing to the Bureau of Alcohol, Tobacco, Firearms and Explosives by a law enforcement agency, resulting from seizure as well as abandonment, buy-back
program, or other recovery method. Moreover, only seized firearms submitted for tracing are included. Firearms submitted for tracing after recovery do
not represent the entire set of all seized firearms.
Source: UNODC IAFQ.
1
2
2
1
64
268
427
1
4
2
2
9
12
244
75
42
104
17
42
124
141
7
11
Air soft gun
Pellet guns
Pellet guns
Air rifle
Air rifles
Air guns²
Air guns²
Air guns
Air guns
Air guns
Air guns
Air rifles, gas pistols
Air rifles, gas pistols
Blank-firing guns and air guns
Blank-firing guns and air guns
Gas pistols
Gas pistols
Gas pistols (air spring), air pistol, air rifles
Gas revolver, gas pistols (air spring), air pistols, air rifles
Air weapons, gas guns, signal pistol
Air weapons, gas guns
Tear gas launchers
Tear gas launchers
0.51248163264128256512
2016
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Antigua
and
BarbudaBotswanaHungaryJapan²MauritiusNepal
North
MacedoniaNorwaySerbiaRomania
United
Kingdom³
United
States of
ŵĞƌŝĐĂϺ
Number of weapons seized (log scale)
0
43
Understanding the black market for firearms 2
of Portuguese citizens with sufficient knowledge and
equipment to modify or otherwise adapt weapons in their
homes. Portugal also reported some seizures of Flobert
guns which entered Portugal illegally from Slovakia.
Among the types of trafficked weapons, Spain identified
“unused” reactivated pistols, blank-firing pistols (“pistolas
detonadoras”) originating mainly from Turkey, “acoustic-
expansion” weapons originating mainly from Czechia and
Slovakia, and homemade weapons. The blank-firing pistols
and “acoustic-expansion” weapons were converted by
making the necessary adaptations in the barrel (eliminat-
ing the stopper or seal) and “unused” pistols were reacti-
vated in order to be able to produce real fire. Homemade
weapons ranged from simple and rudimentary weapons,
such as the so-called "chimbas”, to more sophisticated and
reliable pen guns.
In one jurisdiction in Australia where blank-firing firearms
were sold legally, there was an apparent market for home-
made, modified or converted firearms including shotguns
and blank-firing firearms. There is no information on what
triggered this dynamic, but access to modification instruc-
tions on the internet may have played a role. Another
jurisdiction in Australia detected proscribed modifications
such as sawn-off barrels in the illicit market.
Evidence on diversion
The concept of tracing lies at the heart of efforts to identify
the point of diversion, understood as the point in which
firearms exit the legal circuit and enter the illicit realm. The
United Nations Firearms Protocol, in its Article 3 (f),
defines “tracing” as the “systematic tracking of firearms
and, where possible, their parts and components and
ammunition from manufacturer to purchaser for the pur-
pose of assisting the competent authorities of States Parties
in detecting, investigating and analysing illicit manufactur-
ing and illicit trafficking.” In practice, this may involve a
variety of strategies, depending on national context, to track
the lifecycle of a firearm, but a systematic approach usually
entails an examination of the legally required markings on
the firearm (which are intended precisely for the purpose
of tracing), a check against national records and interna-
tional databases and, where necessary, the submission of
tracing requests to counterparts in other countries.
Such efforts may yield a variety of outcomes, and some
information on these tracing outcomes was collected
through the UNODC questionnaire, in a format designed
to cater for the purposes of SDG indicator 16.4.2.
14
Although these data are geared towards the measurement
of the success rate in tracing, they can also give some infor-
mation about the nature of diversion and the modality
whereby the weapons become illicit.
14 See Section on SDG Indicator 16.4.2 and tracing.
A central question is the extent to which the firearms of
illicit origin
15
in a country’s black market derive from illicit
domestic sources (domestic diversion or domestic illicit
manufacture) as opposed to trafficking from abroad. Some
information on this may be gleaned from data on tracing
outcome, which includes information on whether the fire-
arm was identifiable through marking, whether the firearm
was seized from its legitimate owner,
16
whether a tracing
attempt was made and, if so, whether the firearm could
be traced to a local or foreign registry (see Figure 11).
The above question can be addressed by considering three
different sets of arms of illicit origin, defined in terms of
the tracing outcome. The largest of the three sets consists
of all seized firearms which were uniquely marked (except
for those seized from their legitimate owner). This set cap-
tures the biggest universe of seized firearms which could
provide information on the nature of diversion (domestic
or not), but it carries most uncertainty, as it includes sev-
eral tracing outcomes which do not conclusively determine
whether the illicit origin was of a domestic or transnational
nature. These data show that, on average, illicit origin of
a domestic nature accounted for between 32 to 92 per cent
of all seized arms of illicit origin (see bar “A-F” in Figure
12).
When focusing on the smaller subsets of seized arms, those
for which a tracing attempt was made and those for which
there was successful tracing, this range becomes smaller
and more revealing, although the numbers may be less
representative (since the universe of firearms considered
is smaller – see bars “B-F” and “D-F” in Figure 12).
The smallest set consists of successfully traced arms. In
principle, it can reveal with most precision the proportion
of domestically diverted arms, through data on firearms
that have been traced to a foreign registry (and quite likely
diverted abroad) and firearms that have been traced to the
national registry (and quite likely diverted domestically).
17
These data suggest a share of domestically diverted arms
of around 70 per cent among seized arms of illicit origin.
However, these data provide a partial picture because they
describe only those seizures where it was possible to deter-
mine the point of diversion through successful tracing.
The picture which emerges overall, though not definite,
does suggest that, on average, a significant proportion
(potentially a majority) of arms of illicit origin derive from
domestic sources, at least on the basis of seized arms which
15 See next Section “SDG Indicator 16.4.2 and tracing” for a discussion
of the “illicit origin” of firearms.
16
Firearms held by their legitimate owner cannot be said to be of “illicit
origin” and the question of domestic versus transnational illicit origin
is not applicable.
17
This assumes that the tracing process follows through to the last
legal record of the firearm, implying that the firearm was not legally
exported at a later stage (thus precluding the rather unusual, albeit also
documented, scenario of firearms which are exported, subsequently
diverted and then trafficked back into the country of seizure).
44
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
are identifiable through marking.
18
Although this reflects
the typical scenario to be expected on average, the reality
in a given country will of course depend on its specificities
and the nature of its licit and illicit market. It stands to
reason that, if a given country’s licit market provides an
ample, varied and easily accessible pool of weapons, it may
be more viable for criminals to divert firearms internally
rather than resort to firearms trafficked from other coun-
tries. This holds not only for countries with legal domestic
manufacture but also for countries whose licit market is
mainly supplied by legal imports.
Also, in countries with relatively restrictive laws on the licit
market which severely curtail the range of accessible fire-
arms and the ease of obtaining them, domestic diversion
remains a predominant source for illicit firearms. An exam-
ple is the United Kingdom. Based on the information from
the National Ballistics Intelligence Service (NABIS) of the
United Kingdom, the number of firearms seized there from
legitimate owners is negligible. Moreover, using serial num-
bers and manufacture markings when present, as well as
information on legal sales for those firearms of an age
where sales records were believed to be in existence, NABIS
18 In general, illicit domestic sources could in principle consist of diver-
sion as well as illicit manufacture. For seized arms which are identifi-
able through marking, the contribution of illicit manufacture is less
relevant.
are able to determine the country of diversion of approxi-
mately one third of firearms recovered in the United King-
dom. Analysis of those firearms has consistently shown
that approximately 70 per cent transitioned from lawful
to unlawful possession in the United Kingdom.
It may be the case that, in such countries, some of the
novel and unconventional ways of illicitly manufacturing
a firearm, including assembly, conversion, extensive modi-
fication and reactivation, provide an alternative to traf-
ficking from abroad as a method to supply the illicit
market with weapons that are not available on the licit
market.
Other factors to consider are the levels of legally registered
firearms in a country. The National Firearm Trace Program
of the Australian Criminal Intelligence Commission, for
example, found that, based on firearms traced in 2015-16,
the primary contemporary method of diversion in Aus-
tralia was theft from licensed individuals or firearm dealers
(including suspected staged theft), while illegal import
accounted for a relatively small percentage of illicit fire-
arms in the Australian market.
19
19 Australian Criminal Intelligence Commission, Illicit Firearms in Aus-
tralia, 2016.
FIG. 11
Average breakdown of seized arms by
tracing outcome in a single country,*
2016-17
* Simple averages for 14 countries.
Notes: Black border denotes arms which were uniquely marked. Dotted back-
ground denotes arms not considered to have an “illicit origin”. Striped shading
denotes outcomes which do not provide evidence on the domestic or transna-
tional nature of the illicit origin of an arm (including segment E).
Source: UNODC IAFQ.
FIG. 12
Domestically diverted arms as proportion* of
seized arms of illicit origin – ranges (based on
uniquely marked arms only), 2016-17
* Simple averages.
Note: Letters refer to labelled segments in Figure 11.
Source: UNODC IAFQ.
Seized from
legitimate owner
and found in
national registry
Seized from illegitimate
owner and traced
domestically in national
registry
Point of diversion
otherwise
established by a
competent
authority
Traced
internationally to
foreign registry
Tracing attempted, but not
enough information to identify
point of diversion
Tracing pending
No tracing
procedure
initiated
Not identifiable
through
marking
Unknown status
with respect to
marking
Unaccounted for
E
D
C
B
F
A
0%
20%
40%
60%
80%
100%
Based on all
uniquely marked
arms of illicit origin
Based on tracing
attempts of marked
arms
Based on
successfully
traced arms
Proportion (percentage)
Lower end of range
Range
Upper end of range
A, B, C, D, E, F
B, C, D, E, F
D, E, F
45
SDG indicator 16.4.2 and tracing 2
SDG indicator 16.4.2 and
tracing
SDG indicator
With the adoption of target 16.4 of the 2030 Agenda for
Sustainable Development, countries committed to “by
2030, significantly reduce illicit financial and arms flows,
strengthen the recovery and return of stolen assets and combat
all forms of organized crime”. To measure the attainment
of this target, Member States adopted Sustainable Devel-
opment Goal (SDG) indicator 16.4.2, which measures
the reduction of illicit arms flows by collecting data on the
“proportion of seized, found and surrendered arms whose
illicit origin or context has been traced or established by a
competent authority in line with international
instruments”.
The purpose of the indicator is to establish the illicit origin
of relevant seized, found and surrendered arms, which
entails identifying the specifics of the moment or situation
in which the arm began its illicit lifecycle. For legally man-
ufactured arms, this boils down to establishing the point
of diversion, that is, the moment when the (originally
legal) arms crossed from legal to illegal status. For arms
which were manufactured illicitly, establishing the “illicit
origin or context” requires identification of the circum-
stances in which they came into existence.
This chapter describes the process of disentangling infor-
mation on firearms seizures to compute the SDG indicator
and presents the first global baseline data for the indicator
although still based on data from a relatively small number
of countries. This first data set shows that the practice of
tracing, in order to establish the illicit origin or illicit con-
text of seized firearms—along with the systematic record-
ing of the outcome of tracing—is not very well established.
On average, it was possible to have successful tracing out-
comes for only 28 per cent of the relevant category of
firearms – mainly those with identifying markings.
The countries that achieved high success rates in tracing
firearms reported relatively low seizure levels, perhaps
because tracing requires separate investigative efforts, time
and human resources that are often not available. On the
other hand, some countries with high levels of seizures
registered a low success rate, which may also be linked to
firearms seized in connection with less serious offences
which are not prioritised in tracing.
A pathway through seizure data to
determine success in tracing outcomes
Establishing how many seized firearms could successfully
be traced requires a careful analysis of the data on seized
firearms in all their components. Firstly, it should be noted
that some firearms may be seized from their legitimate
owner, in which case their origin cannot be described as
illicit, and they are not linked to illicit flows. These seizures
are not taken into account when determining if there was
successful tracing and can be removed from the universe
of the indicator, which thus includes in principle any situ-
ation in which the firearm is seized, found or surrendered
in circumstances other than from/by its legitimate holder.
In practice, the process for establishing the illicit origin of
firearms is heavily impacted by the presence of markings.
Industrially manufactured firearms carry identifying mark-
ings which render them uniquely identifiable and enable
authorities to trace their legal lifecycle, and hence the last
legal record and the point of diversion. However, some
firearms are also manufactured in artisanal settings or come
into existence in other illicit ways, in which case markings
may not be present. Some firearms may also have their
markings erased or altered in a way which renders them
not uniquely identifiable any more; such action in itself
can be considered to represent the beginning of the illicit
lifecycle of the firearm.
Thus, the process of tracing the illicit origin of a seized
firearm starts by checking whether the firearm is uniquely
identifiable through markings. This is, however, only the
first step. For uniquely marked firearms, the tracing pro-
cess usually starts by ascertaining whether the firearm in
question is recorded in the national registry of the seizing
country; aside from firearms seized from their legitimate
owner, such a search may yield “successful” outcomes if
the firearm was recorded as “lost”, “stolen”, “confiscated”,
“marked for destruction”, “deactivated”, et cetera. If the
firearm is not recorded in a national registry, the next step
is usually to attempt to identify the last legal record of the
firearm through searches in international databases, such
as Interpol’s iARMS database of stolen, lost and found
firearms, or, if necessary, to locate a legal record of the
firearm in a foreign registry by sending individual tracing
requests through bilateral contacts or other established
tracing mechanisms. The markings may assist in identify-
ing the countries to which such requests may be directed.
Such countries can be the country of manufacture or, if
the relevant markings exist, the country of last legal impor-
tation. If successful, such requests may yield outcomes
similar to the case of items found in national registries.
One helpful outcome for both domestic and international
tracing is if the item is recorded as “exported”, because this
could facilitate tracing back the history of the firearm. In
order to truly identify the point of diversion, it is impor-
tant that the process identifies the last legal record of the
firearm.
For items which were not uniquely identifiable through
marking, the process would entail pursuing a variety of
potential avenues as the opportunity arises, including fur-
ther investigations and intelligence gathering, ballistics
work, communication with international counterparts and
other measures.
46
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
See Figure 13 for a general illustration of the process of
establishment of illicit origin/context of seized firearms.
It should be borne in mind, however, that, due to consid-
erations of data availability as well as the focus of SDG
target 16.4 on illicit flows, this schematic representation
goes beyond the categories used for the computation of
indicator 16.4.2.
The computation of the SDG indicator 16.4.2 requires
to distinguish between “successful” and “unsuccessful”
efforts to establish the illicit origin or context, among the
entire universe of seized arms
20
where this concept is rel-
evant and applicable (that is, excluding arms seized from
their legitimate owner). UNODC’s IAFQ questionnaire
asks Member States to classify the outcome of these efforts
into a number of categories, which are then used to deter-
mine the numbers of firearms for which these efforts were
successful or not.
The data collected by UNODC distinguish between those
firearms which were uniquely identifiable through mark-
ings and those which were not. For those firearms which
20 The SDG Indicator also refers to found and surrendered arms; in view
of the availability of data, the computations so far have been restricted
to seized arms.
were not identifiable through markings, which include
illicitly manufactured firearms and firearms with erased
or altered markings, there is a certain degree of subjectiv-
ity in determining whether the circumstances of the illicit
origin were ascertained with sufficient detail; moreover,
the available data do not allow to distinguish between suc-
cessful and unsuccessful outcomes. For this reason, the
computation of the SDG indicator focuses on the sub-
universe of “potentially traceable” firearms, which consists
mainly of firearms with uniquely identifiable markings
(and also excludes firearms seized from their legitimate
owners). Firearms whose marking status was not recorded
are also included and considered as “unsuccessful” instances
of the efforts to identify the illicit origin. See Figure 14
for an illustration of the computation of the SDG and the
various categories.
Findings on the SDG indicator
Sufficiently detailed data to compute the SDG indicator
were provided by 14 countries. On average, the value of
the indicator amounted to 28 per cent.
21
This number was
21 The average proportion of successfully traced arms is adjusted for arms
which do not belong to the universe of potentially traceable arms. See
Methodological Annex for details.
FIG. 13
Typical process for the establishment of illicit origin/context of seized, found and surren-
dered arms, their parts and components and ammunition
SEIZED, FOUND AND SURRENDERED ARMS
No action taken yet
Actions taken after the apprehension:
i
d
en
t
if
icat
io
n
, t
r
acin
g
,
fu
r
t
h
e
r
i
n
v
estiga
t
i
o
ns ..
.
Item idenƟĮĞd
in f
oreign
registry
Item idenƟĮĞd
in n
ati
on
al
r
e
gi
s
t
ry
Tracing procedure
of checking foreign
registries pending /
not concluded
Item no
t
r
e
c
o
r
de
d
in
ŶĂƟŽŶĂů registry and
n
o
foreign registries
checked
No action taken yet
Actions taken after the apprehension:
f
urther investigations
Illicit use
by its
ůĞŐŝƟŵĂƚe
owner
Conclusive
inforŵĂƟon
through
intelligence
No
conclusive
inforŵĂƟon
on origin
Item illicitly
manufactured
(no original marking)
Illicit
manufacture
idenƟĮĞd
Illicit
manufacture
not idenƟĮĞd
Item with erased or
altered marking
Circumstances
for the
altering of
marking
idenƟĮĞd
Circumstances
for the
altering of
marking not
idenƟĮĞd
ACTION
OUTCOMEORIGIN ESTABLISHED
O
r
igi
n
es
ta
bli
sh
e
d
No conclusive information on illicit origin
I
T
EM UNIQ
UE
LY
I
D
EN
TI
F
IA
BL
E
ITEM NO
T
UNIQUE
L
Y ID
E
NTI
F
I
AB
LE
Items
recorded
as “lost”or
“stolen”
Items
recorded as
͞ĚĞĂĐƟǀĂƚĞĚ͟
Items
recorded as
“marked for
destrucƟon”
Items
recorded as
͞ĐŽŶĮƐĐĂƚĞĚ͟
Items
recorded as
“exported”
Last legal
owner
idenƟĮĞd
FIG. 14
Computation of SDG Indicator 16.4.2
=
Seized from
illegitimate owner
and traced domesti-
cally in national
registry
Traced
internationally to
foreign registry
Point of diversion
otherwise
established by a
competent
authority
Seized from
illegitimate owner
and traced domes-
tially in national
registry
Traced
internationally to
foreign registry
Point of diversion
otherwise
established by a
competent
authority
Tracing attempted,
but not enough
information to
identify point of
diversion
Tracing pending
No tracing
procedure initiated
Unknown status
with respect to
marking
SDG 16.4.2
47
SDG indicator 16.4.2 and tracing 2
driven by two components in roughly equal measure: fire-
arms traced domestically to a national registry (13 per
cent) and firearms traced internationally to a foreign reg-
istry (15 per cent).
There was notable variability in the value of SDG indica-
tor 16.4.2 across the 14 countries. In particular, some very
high proportions of traced firearms were registered in some
countries with relatively low levels of seized arms, such as
Bahamas (average of 362 arms seized per year in 2016-17)
and Azerbaijan (215 per year). At the other extreme, some
very low proportions were registered by countries with
relatively high numbers of arms seized, such as Kenya
(9,728 firearms seized in 2017) and Australia (average of
26,660 arms seized per year in 2016-17).
This apparent association between high seizure levels and
low values for the SDG indicator could potentially reflect
two different mechanisms. Smaller quantities of firearms
to be traced create less of a burden for tracing which is
more likely to be carried out successfully (and conversely
higher numbers, a bigger burden, and less successful trac-
ing). Higher seizure levels may also be indicative of firearms
seized for less serious offences, which may in turn not be
deemed to warrant the investment of resources dedicated
to tracing. This last inverse association is mainly observ-
able at the extreme values of the indicator and cannot be
said to constitute a general pattern (see Figure 16).
A breakdown of the different types of tracing outcomes
gives further insights into the composition of the success-
ful outcomes. In some countries, such as Argentina, Aus-
tralia, Azerbaijan, Peru, Republic of Moldova and Spain,
the dominant contribution was the component of firearms
found in a national registry. In general, such a pattern is
not surprising, given that tracing to a foreign registry is
FIG. 15
Typical breakdown* of seizures by tracing outcome and SDG Indicator 16.4.2, 2016-17
* Average of 14 countries.
Source: UNODC IAFQ.
inherently more onerous in comparison with tracing to a
national registry.
In some cases, such as Argentina, the predominance of
domestic tracing may be driven by a significant compo
-
nent of seized firearms which were manufactured in the
Seized from legitimate
owner and found in
national registry
Not identifiable
through marking
Unaccounted for
No tracing
procedure
initiated, 34%
Tracing attempted, but not enough
information to identify point of
diversion, 11%
Point of diversion otherwise established by a
competent authority, 0.6%
Seized from illegitimate owner and traced
domestically in national registry, 13%
Traced internationally to foreign
registry, 15%
Unknown status with respect to
marking, 10%
Tracing pending, 17%
Potentially
traceable
arms of
illicit origin
FIG. 16
Value of SDG Indicator 16.4.2, in comparison
with average arms seized, 2016-17
1
Data for 2017 only.
2
Include seizures of an administrative nature.
3
Data for 2016 only.
4
Both the average number of arms seized as well as the value of the SDG Indicator are
based only on seizures in the records of the Federal Police Tracing Centre of Brazil.
Source: UNODC IAFQ.
0%
50%
100%
0
5,000
10,000
15,000
20,000
25,000
30,000
Kenya¹
Romania
Australia²
Lithuania³
Antigua and Barbuda
Republic of Moldova
ƌĂnjŝůϺ
United Kingdom
Argentina
Spain
Grenada
Peru
Bahamas
Azerbaijan
SDG Indicator 16.4.2 (percentage)
Average number of arms seized
SDG Indicator 16.4.2, 2016-17 (right axis)
Average arms seized per year, 2016-17 (left axis)
48
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
country and diverted domestically (at least 69 per cent of
firearms seized in Argentina in 2016-17 had been manu-
factured in Argentina itself
22
).
In some cases, the lack of successful international tracing
to a foreign registry was due to countries’ limited capacity
to report on this number. However, some countries also
reported that no firearms were traced internationally, sug-
gesting that greater capacity and awareness raising are
needed in this area.
In other countries, such as Bahamas, Brazil and the United
Kingdom, the dominant component of successful tracing
outcomes was that of firearms traced to foreign registries.
It should however be borne in mind that these computa-
tions are based on a subset of all seized firearms.
Tracing requests sent and received
Aside from the outcome of tracing efforts, countries were
also asked to provide information about the number of
tracing requests sent during 2016-17, and to which coun-
tries or agencies these requests were addressed. In total, 15
22 Out of 43,321 arms seized in 2016-17, 29,794 had been manufac-
tured in Argentina; in addition, the country of manufacture was
unknown for 4,251 arms.
FIG. 17
Breakdown of potentially traceable arms by tracing outcome, and composition of
Sustainable Development Indicator 16.4.2 by type of successful outcome, 2016-17
* Based on data for 2017 only.
** Based on data for 2016 only.
Note: Successful tracing outcomes are shown on the left hand side in different shades of purple; unsuccessful tracing outcomes are shown on the
right hand side in different shades of brown.
Source: UNODC IAFQ.
SDG Value: 8%
SDG Value: 18%
SDG Value: 1%
SDG Value: 100%
SDG Value: 100%
SDG Value: 14%
SDG Value: 65%
SDG Value: 0%
SDG Value: 2%
SDG Value: 78%
SDG Value: 9%
SDG Value: 0%
SDG Value: 55%
SDG Value: 15%
Antigua and Barbuda
Argentina
Australia
Azerbaijan
Bahamas
Brazil
Grenada
Kenya*
Lithuania**
Peru
Republic of Moldova
Romania
Spain
United Kingdom
Seized from illegitimate owner and found in national registry (successful tracing)
Seized from illegitimate owner and found in foreign registry (successful tracing)
Point of diversion otherwise established (successful tracing)
Tracing attempted, but not enough information to identify point of diversion
Tracing pending
No tracing initiated
Unknown status with respect to marking
0% 20% 40% 60% 80% 100%
countries reported having sent 6,173 requests to 45 dif-
ferent countries and one agency (Interpol) (see Figure 18).
The largest numbers of sent requests were reported by
countries in Latin America and the Caribbean; in some
cases, similar requests pertaining to the same firearms may
have been sent to multiple countries in a systematic fash-
ion. In terms of the number of different countries to which
requests were sent, the highest numbers were reported by
the United Kingdom (26) and Spain (13); however, based
on a small subset of tracing requests for which details were
available (23 out of 837 made during the period 20 Feb-
ruary 2017-31 December 2017), Brazil sent tracing
requests to at least 15 different countries.
The United States was most frequently mentioned as the
country to which tracing requests were sent (mentioned
by 6 countries), followed by Germany (4 countries). In
addition, 7 countries reported sending requests to, or facil-
itated by, Interpol.
There appeared to be a certain correspondence between
the country of manufacture of seized firearms and the
countries to which tracing requests were sent. For example,
in the case of the United Kingdom, among the top nine
countries which were identified as the country of manu-
49
SDG indicator 16.4.2 and tracing 2
FIG. 18
Tracing requests sent, by reporting
country, 2016-17
1
For Costa Rica, El Salvador and Republic of Moldova the total number
of arms involved was not available.
2
Data for Brazil refer to the period 20 February 2017-31 December
2017 only. The total number of arms involved and the total number of
countries to which requests were sent were not available.
3
For Italy and Republic of Moldova, all requests were sent to or through
Interpol.
4
Data from one jurisdiction only.
5
Data for Costa Rica and the United Kingdom refer to 2017 only.
Source: UNODC IAFQ.
FIG. 19
Tracing requests sent by the United Kingdom, by
requested country, in comparison with country
of manufacture of arms seized by the United
Kingdom, 2017
* Top 9 countries, excluding the United Kingdom itself and unknown countries of
manufacture. Other countries included under "Other".
** Top 12 countries to which requests were sent. Other countries included under
"Other".
Source: UNODC IAFQ.
manufacture may also play a role in the cases of Italy, Spain
and the United Kingdom—among countries which pro-
vided data on tracing requests received, Italy, Spain, Bel-
gium and the United Kingdom constitute the top four
counties in terms of their frequency as the country of
manufacture of seized firearms (compare Figure 20 and
Figure 24).
In some cases, the high numbers of requests appeared to
be driven by requests between neighbouring countries; for
example, El Salvador reported sending a total of more than
of 4,000 requests to Mexico, Guatemala, Honduras, Nica-
ragua and Costa Rica, and receiving a total of more than
3,500 requests from Mexico, Honduras and Costa Rica.
A similar pattern held also for Costa Rica, which sent
requests to eight countries, of which seven were in Central
America, and identified only Panama among the countries
from which it received tracing requests.
In other cases, the high numbers of requests were received
from a very diverse group of countries. Brazil, Spain and
the United Kingdom, which sent tracing requests to the
highest numbers of countries, were also among those
reporting requests received from the highest numbers of
countries: at least 20 in the case of the United Kingdom,
at least 13 in the case of Spain (aside from Europol and
Interpol) and 12 in the case of Brazil. This suggests that a
0
1
10
100
1,000
10,000
Peru
Slovenia
Albania
Republic of Moldova¹,³
Slovakia
Grenada
Romania
Italy³
ƵƐƚƌĂůŝĂϺ
Spain
hŶŝƚĞĚ<ŝŶŐĚŽŵϻ
ŽƐƚĂZŝĐĂϷ͕ϻ
Bahamas
Brazil²
El Salvador¹
Number of tracing requests
(logarithmic scale)
Number of requests sent
Total number of arms involved
Number of countries to which requests were sent
facture of firearms seized by the United Kingdom (exclud-
ing the United Kingdom itself), most (seven) were also
among the top nine to which the United Kingdom sent
tracing requests. This is to be expected, as the country of
manufacture is a natural starting point in tracing the life-
cycle of a firearm (see Figure 19).
Countries were also asked to provide information about
the number of tracing requests received during 2016-17,
and from which countries or agencies these requests origi-
nated. In total, 16 countries reported receiving 6,915
requests from 65 countries and 2 agencies (Europol and
Interpol) (see Figure 20).
Some of the countries which reported sending high num-
bers of tracing requests were also among those which
received relatively high numbers of requests; this included
Costa Rica, El Salvador, Spain as well as the United King-
dom (in view of the fact that data available for this country
covered 2017 only).
Belgium attributed the large number of tracing requests
received (769 from at least 39 countries) to the presence
of a large firearms production facility in Belgium. Local
Turkey
Czechia
France
Italy
Germany
Belgium
Brazil
Russian
Federation
Spain
Japan
Serbia
Slovakia
Other
Turkey
France
Italy
Germany
Belgium
Russian
Federation
Spain
Japan
United States
of America
Other
Country of
manufacture of
seized firearms*
50
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 21
Tracing requests sent to, received
from, or otherwise facilitated by
Interpol, 2016-17
1
Includes requests placed through the Interpol platform iARMS.
2
Brazil reported sending 18 requests involving Interpol; however details
were unavailable for the vast majority of requests and therefore this
number likely underestimates the actual figure.
3
Spain also reported 12 requests coming from Europol.
Note: Slovakia reported 381 incoming requests, of which an unspecified
number were received from Europol and Interpol.
Source: UNODC IAFQ.
general engagement of such countries in the area of trac-
ing begets mutually reinforcing roles of receiving and
responding to tracing requests.
Nine countries reported the involvement of Interpol in
tracing requests sent or received (see Figure 21). Among
these countries, the number of tracing requests received
tended to be higher than requests made; this may be influ-
enced by the ability of countries to effectively direct trac-
ing requests to multiple countries through the use of a
dedicated multilateral platform managed by Interpol.
Licit manufacture
It is important to note that, while the illicit supply chain
starts at the point of diversion (or illicit manufacture), a
diverted weapon may have a long licit history prior to
diversion, starting with the country of licit manufacture.
With the exception of some countries with significant
domestic licit manufacture of firearms, typically the major-
ity of seized weapons were manufactured outside the coun-
try of seizure. This is hardly surprising, given that, in many
countries, the licit market also relies heavily on imports
from the major producing countries (see Figure 22).
The country of diversion may be anywhere on the chain
between the country of manufacture and the country of
seizure. The indications that the illicit source of firearms
is often to be found in the country where firearms are
seized (see Figure 12), coupled with the fact that seized
arms are, for most countries, manufactured outside the
country of seizure, suggest that the country of manufac-
ture may have little to do with the country of diversion.
There is also the possibility of undetected diversion hap-
pening in manufacturing countries, but this could not be
ascertained with the available information.
Further corroboration of this can be found when consid-
ering a special subset of seized arms which by definition
are unlikely
23
to have been diverted in the country of sei-
zure - namely arms seized on incoming cross-border ship-
ments. Among such seizures, there appears to be a
significant proportion which enter from a country other
than the country of manufacture. This does not, by itself,
establish what proportion of arms are diverted within the
23 Aside from the possibility that a firearm was diverted in a given
country, then trafficked into another country, followed by an attempt,
detected at the border, to traffic the firearm back into the country of
diversion.
FIG. 20
Tracing requests received, by report-
ing country, 2016-17
1
Data from one national agency only.
2
For Belgium, Costa Rica, El Salvador, Republic of Moldova and Slovakia
the total number of firearms involved was not available.
3
For Costa Rica and Slovakia, the number of requesting countries was
not available.
4
Spain received requests from 13 countries or more.
5
For Italy, all requests were received from or through Interpol.
6
For Belgium, the number of requesting countries may omit an
unknown number of countries reported under "other".
7
Data for Belgium and the United Kingdom refer to 2017 only.
8
For the United Kingdom, the number of requesting countries may omit
one country reported under a regional aggregation.
Source: UNODC IAFQ.
0
1
10
100
1,000
10,000
Australia¹
Peru
Libya
Tunisia
Brazil
Romania
Republic of Moldova²
Slovenia
hŶŝƚĞĚ<ŝŶŐĚŽŵϼ͕Ͻ͕Ͼ
Albania
^ƉĂŝŶϹ͕Ϻ
Slovakia²,³
/ƚĂůLJϻ
ĞůŐŝƵŵϸ͕ϼ͕Ͻ
Costa Rica²,³
El Salvador²
Number of tracing requests (logarithmic scale)
Number of requests received
Total number of arms involved
Number of requesting countries
1
10
100
1,000
10,000
Albania¹
Belgium
Brazil²
Costa Rica¹
Italy
Republic of Moldova
Romania
Slovenia¹
Spain³
Number of requests (logarithmic scale)
Reporting
country
Outgoing requests
Incoming requests
51
Licit manufacture 2
an exportation to another country – unless the owner her-
self moves across the border (temporarily or to take up
residence for the longer term
24
). Thus, for firearms in civil-
ian possession, the country of the last legal status may
frequently coincide with the country of the first retail pur-
chase, with individual mobility being the main counter-
balance to this. In other words, if an illicit firearm was at
some point diverted from civilian ownership, it is quite
plausible that this happened either within the country of
first retail purchase, or else in an outgoing cross-border
movement from the country of first retail purchase.
25
The concept of tracing
26
requires the identification of the
point at which the transition is made from the licit to the
illicit spheres (point of diversion). In principle, this
requires the identification of the last legal owner; in some
countries where there is no legal obligation to keep records
of the final end-user, this information is not possible to
obtain and tracing practices will by necessity look for the
retail purchase in order to determine with some approxi-
mation the point of diversion.
24 For example, sports shooters and hunters may cross borders with their
legally held firearms and sell their arms during their travel.
25
One exception to this is the possibility of legal exports which are really
pretexts for diverting the weapon in the country of importation.
26
See Article 3 of the UN Firearms Protocol and paragraph 5 of the
International Instrument to Enable States to Identify and Trace, in a
Timely and Reliable Manner, Illicit Small Arms and Light Weapons.
FIG. 22
Proportion of seized arms manufac-
tured outside country of seizure, by
country, 2016-17
Note: Proportions are adjusted to exclude arms whose country of manu-
facture was unknown or not reported.
* These reporting countries did not indicate themselves among the coun-
tries of manufacture, but reported some seized arms manufactured in
unspecified "Other" countries (a breakdown by country was only col-
lected for the main (top 10) countries of manufacture.)
** Based on data for revolvers and pistols only (data on country of man-
ufacture for overall seizures were not available).
*** Based on data for 2016 only. The response from Burkina Faso sug-
gests that arms manufactured in Burkina Faso itself (using artisanal tech-
niques) accounted for the largest share of arms seized in the country,
but the number of such arms was not quantified. As a proxy, the total
number of artisanally manufactured arms is used in the calculation.
However, some arms manufactured in neighbouring countries may also
have been manufactured using artisanal techniques; hence the propor-
tion manufactured outside Burkina Faso may be under-estimated.
Source: UNODC IAFQ.
FIG. 23
Country of provenance of arms
seized on an incoming route, break-
down into country of manufacture
versus third country, 2016-17
* This distribution is heavily impacted by two large seizures by the same
country and from the same country of provenance. Excluding these two
large cases, the distribution is: 22 per cent of weapons for which the
country of provenance coincided with the country of manufacture, 78
per cent of weapons for which the country of provenance was another
country.
Note: Based on reported significant cases of weapons seized on an
incoming route for which the country of provenance (country from
which the seized arms entered the country of seizure) and as the coun-
try of manufacture of the seized weapons were both available (7 coun-
tries).
Source: UNODC IAFQ.
Burkina Faso***
Argentina
Spain
Brazil
Japan**
Australia
United Kingdom
Sweden
Mexico
Tunisia
Tajikistan*
Romania*
Republic of Moldova*
Qatar
Portugal*
Paraguay*
Nepal
Myanmar*
Luxembourg
Kenya*
Guatemala
Grenada*
Denmark*
Central African Republic
Azerbaijan
Antigua and Barbuda
Algeria*
0%
50%
100%
150%
Country of seizure
Percentage, by country
Average
country of manufacture or otherwise, but it gives a strong
indication that diversion has occurred after exportation
from the country of manufacture (see Figure 23); in other
words, it is very plausible in such a scenario for an arm to
be first legally manufactured, then legally exported, and
only then diverted.
The different distribution between number of cases and
number of arms is affected by two large seizures demon-
strating the attempt to illegally export a large quantity of
arms from a manufacturing country. This suggests that
there may be cases with a large number of trafficked fire-
arms where the point of diversion is in the country of legal
manufacturing (or at its border). However, such cases were
relatively infrequent; once more, this is not surprising,
given that many civilian firearms are exported at wholesale
level and only subsequently sold at retail level outside of
the country of manufacture.
Once a firearm has been legally sold and is in civilian own-
ership, it is rather unlikely for the legitimate owner to
undertake, in a legal fashion and for purely legal purposes,
53%
17%
47%
83%
0% 20% 40% 60% 80% 100%
In terms
of number
of arms*
In terms
of number
of cases
Country of manufacture Other country
52
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
TABLE 1
Most frequently reported countries of manufacture of seized arms, ranked by number
of reporting countriesª (by geographical proximity of reporting country), 2016-17
* In addition, one African country reported seized arms manufactured in the Union of Soviet Socialist Republics (former State).
a
Not including countries reporting seized arms manufactured in their own countries.
Source: UNODC IAFQ.
TABLE 2
Most frequently reported countries of manufacture of seized arms, ranked by number
of reporting countriesª (by region of reporting country), 2016-17
* In addition, one African country reported seized arms manufactured in the Union of Soviet Socialist Republics (former State).
a
Not including countries reporting seized arms manufactured in their own countries.
Source: UNODC IAFQ.
Country of manufacture
Reporting countries, in relation to country of manufacture
Intra-subregionalª Intra-regional Inter-regional Total
Italy 3 5 15 23
Germany 4 5 13 22
United States of America 6 15 21
Belgium 4 3 10 17
Czechia 1 6 9 16
Russian Federation 1 4 10* 15*
Spain 2 5 7 14
Turkey 2 1 11 14
China 2 2 8 12
Austria 2 3 5 10
France 2 3 4 9
Brazil 1 2 5 8
United Kingdom 1 1 2 4
Israel 4 4
Japan 1 2 3
Yugoslavia (former State) 3 3
Country of manufacture
Region of reporting country
Africa Americas Asia
Europe Oceania Total
Italy 37481 23
Germany 35491 22
United States of America 26481 21
Belgium 2537 17
Czechia 14371 16
Russian Federation 5* 1 4 5 15*
Spain 22271 14
Turkey 3236 14
China 23421 12
Austria 5 5 10
France 4 5 9
Brazil 3131 8
United Kingdom 1 2 1 4
Israel 2 2 4
Japan 1 1 1 3
Yugoslavia (former State) 3 3
53
Licit manufacture 2
In general, the country of last legal record and that of first
retail purchase may often coincide, but this relationship
deserves further research.
Countries of manufacture
The vast majority of trafficked firearms originate as legally
manufactured firearms which are at some point diverted
to the black market. The above analysis indicates that fire-
arms may leave the country of manufacture through legal
trade, with diversion happening after the first border cross-
ing. Therefore, the country of manufacture of a trafficked
firearm may in principle be far removed from the traffick-
ing route, and thus does not necessarily coincide with the
starting point of the illicit flow of a trafficked firearm.
Nevertheless, information on the country of manufacture
of seized firearms may yield additional insights into their
provenance. This determination of the country of manu-
facture can be based on verifiable criteria, such as the
markings present on firearms as well as (to a certain extent)
the particular brand or model of the weapon.
27
These
insights can be useful, for example, to improve or put in
place preventive measures at the early stages of the supply
chain.
Based on data on the country of manufacture reported
through the Illicit Arms Flow Questionnaire, several
European countries emerge among the most prominent
countries of manufacture. This is most pronounced in the
case of Germany and Italy, which stood out in terms of
27 The model of a weapon may not necessarily uniquely determine its
manufacturer; however, depending on the particular case, information on
the model and the brand, coupled with a knowledge of the market for fire-
arms, may help to identify or narrow down the country of manufacture.
two different metrics: the number of countries which
identified them among the countries of manufacture of
seized weapons, as well as the proportion of seized arms
originating from the country–in terms of manufacture
28
(see Table 1, Table 2 and Figure 24). The United States
and the Russian Federation also emerge among the most
prominent countries, especially in terms of the second
metric.
At the level of subregional groupings (see Figure 25),
Northern and Western Europe emerged consistently as a
prominent subregion of manufacture across most subre-
gions of seizure. Southern Europe (excluding the Western
Balkans) also accounted (in terms of manufacture) for
significant proportions of arms seized across various sub-
regions, including other parts of Europe but also in the
Americas and within Southern Europe itself. The role of
Northern America as a subregion of manufacture was most
marked in the Americas as well as parts of Asia and Oce-
ania, while arms manufactured in Eastern Europe were
prominent in Africa and Western Asia. South America was
another subregion which seized, on average, a large share
of arms manufactured within the subregion itself, and
specifically within the country of seizure.
A comparison of the location of manufacture of seized
arms with incoming illicit flows of weapons illustrates the
complex relationship that may exist between trafficking
routes and countries of manufacture. In the case of South
America, for example, illicit incoming inter-regional flows
are dominated by the share from Northern America, while
28 These calculations do not include instances of countries reporting
seizures of arms manufactured in their own countries.
FIG. 24
Typical distributionª of country of manufacture of seized arms, 2016-17
a
Simple average over 29 reporting countries
* Includes one aggregate mention for Russian Federation and former Union of Soviet Socialist Republics.
** Includes one aggregate mention for Czechia, Slovakia and former Czechoslovakia.
Source: IAFQ
Unknown/
not available
Country
of seizure
United States
of America
Russian
Federation*
Former USSR
Italy
Germany
Spain
Turkey
Belgium
United Kingdom
Czechia**
China
France
Austria
Former Yugoslavia
22 other countries
Other than
country of
seizure
54
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 25
Breakdown of seized arms by subregionª of manufacture, according to subregionª
of seizure, 2016-17
a
See Methodological Annex for the list and composition of subregional groupings used.
b
Data were not available for all subregions.
c
Percentages are normalized so that the shares of arms manufactured outside the subregion of seizure sum up to one (for each subregion of seizure).
Weapons seized outside the subregion of manufacture are represented proportionally on the left hand side. The calculations adjust for the share of
seized arms whose country of manufacture was not known or not reported.
* Excluding Western Balkans.
Source: UNODC IAFQ.
Europe accounts, in contrast, for the vast majority of seized
weapons in terms of manufacture. Similarly, in the case of
Southern Europe (excluding the Western Balkans), North-
ern America accounts for a large share in terms of countries
of manufacture, but not in terms of incoming illicit flows.
On the other hand, certain similarities between the two
types of distribution can also be seen, especially in the case
of Northern and Western Europe, as well as the role of the
Western Balkans in the case of Southern Europe.
Arms manufactured
insideण subregion
where they were
seized, relaƟve to
other arms
Subregion
a
of seizure࣎
(reporƟng countries)
Subregion
a
of manufacture
DistribuƟon of arms manufactured outsideण subregion
where they were seized (total=100%)
Eastern and
South-Eastern Asia
Eastern
Europe
Northern
America
Northern and
Western Europe
Oceania
South
America
Southern
Europe*
Western Asia
Western Balkans
Caribbean
(2 countries)
Central America
(2 countries)
Eastern and South-Eastern
Asia (2 countries)
Eastern Europe
(2 countries)
Northern Africa
(2 countries)
Northern and Western
Europe (4 countries)
Oceania
(1 country)
South America
(3 countries)
Southern Europe*
(4 countries)
Western and Middle Africa
(2
countries)
Western Asia
(2 countries)
Rest of the world
(3 countries)
Arms manufactured in other country within same subregion as country of seizure
Arms manufactured in country of seizure
90% 0% 60% 0% 40% 0%40% 0% 80% 0% 40% 0%40%0%40% 0% 90% 0%
55
Licit manufacture 2
FIG. 26
Distribution of subregions identified as
departure of incoming illicit flowsª (inner
circle) and manufactureª (external circle)
for seizures made in South America,
2016-17
* Excluding Western Balkans.
a
The largest share of transnational illicit flows affecting countries in South
America occurs between countries within South America. Similarly, countries
in South America itself account for the largest share of manufacture of arms
seized in South America (including arms seized in the country of manufac-
ture). These shares are not shown in the above figure.
Note: The shares of flows and of manufacture are based on different kinds of
data which require different methodologies. Therefore the comparison should
be made with caution. In both cases, the calculations adjust for the share
which is not classified or reported as unknown.
Source: UNODC IAFQ.
FIG. 27
Distribution of subregions identified as
departure of incoming illicit flowsª (inner
circle) and manufactureª (external circle)
for seizures made in Southern Europe,*
2016-17
* Excluding Western Balkans.
a
Southern Europe itself accounts for the largest share of manufacture of
weapons seized in Southern Europe. Similarly, there is also a small share of
transnational illicit flows affecting countries in Southern Europe* which
occurs between countries within Southern Europe. These shares are not
shown in the above figure.
Note: The shares of flows and of manufacture are based on different kinds of
data which require different methodologies. Therefore the comparison should
be made with caution. In both cases, the calculations adjust for the share
which is not classified or reported as unknown.
Source: UNODC IAFQ.
Northern
America
89%
Northern
and
Western
Europe
7%
Southern
Europe*
2%
Western
Asia
2%
Eastern and
South-
Eastern Asia
Eastern
Europe
2%
Northern
America
17%
Northern
and
Western
Europe
10%
Southern
Europe*
66%
Western
Asia
5%
Western
Balkans
Flows
Manufacture
Eastern
Europe
53%
Northern
and
Western
Europe
18%
Western
and Middle
Africa
1%
Western
Asia
18%
Western
Balkans
10%
Eastern and
South-Eastern Asia
1%
Eastern Europe
2%
Northern
America
27%
Northern
and
Western
Europe
53%
South
America
8%
Western
Asia
2%
Western
Balkans
7%
Flows
Manufacture
56
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
FIG. 28
Distribution of subregions identified as departure of incoming illicit flowsª (inner circle)
and manufactureª (external circle) for seizures made in Northern and Western Europe,
2016-17
* Excluding Western Balkans.
a
The largest share of transnational illicit flows affecting countries in Northern and Western Europe occurs between countries within Northern and West-
ern Europe. Similarly, Northern and Western Europe accounts for the largest share of countries of manufacture of weapons seized in Northern and
Western Europe. These shares are not shown in the above figure.
Note: The shares of flows and of manufacture are based on different kinds of data which require different methodologies. Therefore the comparison
should be made with caution. In both cases, the calculations adjust for the share which is not classified or unknown.
Source: UNODC IAFQ.
Eastern and
South-
Eastern Asia
3%
Eastern
Europe
20%
Northern
America
24%
South
America
1%
Southern
Europe*
3%
Western
Asia
37%
Western
Balkans
9%
Rest of
the world
3%
Eastern and
South-Eastern Asia
1%
Eastern
Europe
16%
Northern
America
24%
Southern
Europe*
31%
Western
Asia
16%
Western
Balkans
12%
Flows
Manufacture
57
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
CHAPTER 3
THE NATURE OF
FIREARMS TRAFFICKING
Chapter overview
This chapter analyses the nature of firearms trafficking,
starting by considering the prices paid for firearms in illicit
markets around the world. The illicit nature of firearms
markets makes prices generally higher than for firearms
bought regularly, although there are some exceptions for
some types of firearms in certain geographical locations
where the licit markets compete with lower prices in the
illegal market. The chapter then looks at the different ways
that firearms can be trafficked between different locations,
both across international borders and within countries,
before finally looking more closely at trafficking within
some selected countries with known, sizable domestic fire-
arms trafficking flows.
Due to the considerable risk and costs involved in supply-
ing illicit firearms, in most parts of the world, the illicit
prices of firearms are usually significantly higher than the
licit price. There are, however, some notable exceptions.
In some parts of Latin America and the Caribbean, the
illicit prices were found to be lower than the licit price,
specifically for handguns, which might be due to their easy
availability in the illicit market. Looking at specific fire-
arms types, user preferences also seem to play a role. For
example, in some of these countries, revolvers are report-
edly less expensive in illicit markets than pistols, and the
primary reason appears to be that criminals prefer
pistols.
Trafficking of firearms can be carried out by using differ-
ent strategies, chosen on the basis of the perceived risk of
detection, ease of concealment, feasibility of large payloads
and geographical conditions. For transnational trafficking,
the vigour of border controls along the trafficking route
is another important factor. Regardless of strategy, seizures
that are associated with trafficking tend to involve larger
numbers of firearms per case than those not related to
trafficking (and related to illicit possession, for
example).
Seizures at borders comprise on average less than 10 per
cent of firearms seizures at the national level. In other
words, more than 9 in 10 seizures take place within the
national territory.
1
Even though seizures at the borders are
in the minority, they shed considerable light on transna-
tional illicit flows. For example, they reveal that while the
number of cases involving customs-related seizures from
vessels is relatively low, each case involved nearly 20 seized
weapons. On the other hand, seizures from vehicles
accounted for three quarters of all the customs-related
seizure cases, but fewer than two firearms were seized per
case, on average. There is also a clear distinction between
1 Throughout this report, the term “within national territory”, in refer-
ence to seizures, excludes seizures made at borders. See also glossary.
58
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
the detection of incoming and outgoing flows. Most weap-
ons are seized on an incoming, rather than outgoing, route,
suggesting that the level of scrutiny exercised by customs
authorities tends to be higher with regard to incoming
firearms. It may also be indicative of existing forms of
informal cooperation and information exchange between
countries leading to seizures (and capture of suspects) in
the country of arrival.
It appears that trafficking by land is the most commonly
used mode of firearms trafficking. While information is
scarce, several countries highlighted trafficking by land;
either in road vehicles which may have been modified for
this purpose, in packages or deliveries, or by persons on
foot. Cross-border trafficking of firearms sometimes takes
the form of so-called “ant-trafficking”, by which the illicit
flow involves numerous individuals transporting one or a
small number of firearms in order to lower the risk of
detection. This mode may not be predominant at the
global level, but it might be significant along certain
routes.
Illicit movements of firearms within given countries can
only be gleaned through sub-national-level seizure data.
Such data were available for some countries in Central
America, where it appears that violent crime is a key driver
of demand for firearms. Moreover, seizures are also mark-
edly higher in many border and port areas, as well as in
areas with high levels of trade activity. Sub-national data
from the United States showed that more than 70 per cent
of traced seized weapons had been purchased in the state
where they were seized.
Based on seizures, transnational trafficking flows seem to
be mostly concentrated within continents with the notable
exceptions of inter-regional flows from Northern America
and Europe. Northern America, Europe and Western Asia
together accounted for almost all departure points of traf-
ficking in 2016-17, while Central and South America
together with West Asia accounted for more than 80 per
cent of the trafficking destinations.
Prices of firearms in illicit markets
Further insight into firearms trafficking can be obtained
from prices of firearms on the black market, as well as the
way these relate to licit prices. Some countries provided
prices of the same, or comparable, types of firearm in both
the licit and illicit markets, and also - where possible – on
distinguishing features, such as condition or model.
In most cases, the firearm price follows what happens in
other regulatory markets: prices are distinctly higher in the
illegal market than in the legal market. There are, however,
remarkable exceptions in some geographical locations with
firearm prices lower in the illegal market, suggesting that
in these locations either the illicit supply is much higher
FIG. 1
Ratio of illicit to licit prices of firearms, by country, 2016-17
a
A ratio larger than 1 means that the price on the illicit market was higher than the corresponding price (for the same type of firearm) than the price
on the licit market. A ratio smaller than 1 means that the price on the illicit market was lower.
b
For North Macedonia, the same model of rifle was not available in the licit market; the comparison is based on a similar model.
c
For Australia, Dominican Republic, North Macedonia, Saint Lucia and Tunisia the ratio is based on the midpoint of the reported price ranges.
Note: Only data where price was available for comparable models of firearms in both the licit and illicit markets are included.
Source: UNODC IAFQ.
Pistol
ZŝŇĞ
ZĞǀŽůǀĞƌ
Shotgun
DŽĚŝĮĞĚďůĂŶŬĮƌŝŶŐĮƌĞĂƌŵ
WŝƐƚŽů͕͘ϰϬĐĂůŝďƌĞ
WŝƐƚŽů͕ϵŵŵĐĂůŝďƌĞ
ƐƐĂƵůƚƌŝŇĞ
ZŝŇĞƉĂƌƚŬŝƚƐ
ZĞǀŽůǀĞƌ
Pistol
Pistol
ZĞǀŽůǀĞƌ
Shotgun
Pistol
Pistol
ZŝŇĞ࣎
Pistol
ZŝŇĞ
ZĞǀŽůǀĞƌ
Shotgun
^ƵďŵĂĐŚŝŶĞŐƵŶ
Pistol
ZŝŇĞ
WŝƐƚŽů;ŵŽĚĞůϭͿ
^ƵďŵĂĐŚŝŶĞŐƵŶ
ZŝŇĞ
ϭ
ϭϬ
Ϭ͘ϭ
ϭ
ϭϬ
ƵƐƚƌĂůŝĂण ƌĂnjŝů ŽŵŝŶŝĐĂŶ
ZĞƉƵďůŝĐण
El
^ĂůǀĂĚŽƌ
>ŝƚŚƵĂŶŝĂ EŽƌƚŚ
DĂĐĞĚŽŶŝĂण
^ĂŝŶƚ
>ƵĐŝĂण
^ůŽǀĂŬŝĂ ^ƉĂŝŶ dƵŶŝƐŝĂण
ZĂƟŽ
a
;ůŽŐĂƌŝƚŚŵŝĐƐĐĂůĞͿ
RaƟŽ> 1
IlliĐitƉƌiĐĞhighĞƌthĂnliĐitƉƌiĐĞ
RaƟŽф 1
IlliĐitƉƌiĐĞůŽǁĞƌthĂnliĐitƉƌiĐĞ
59
The nature of firearms trafficking 3
than the licit supply, or that there is not enough interdic-
tion to raise the price in the illegal supply-chain. One
example of these exceptions is the low price of certain types
of firearm, mainly handguns, in the illegal market in Latin
America and the Caribbean. In Brazil, this consisted mainly
of .32 and .38 calibre revolvers manufactured between
1960 and 1997 (the year of the first law on firearms in
Brazil) as well as 9mm and .40 calibre pistols. Brazil
explained the relatively low prices in terms of the ample
supply. The revolvers reportedly originated partly in Brazil
and partly in the United States, while pistols were mainly
trafficked into Brazil through Paraguay, following impor-
tation from several countries. In May 2018, imports of
arms and ammunition into Paraguay were suspended.
2
The higher comparative price of assault rifles in Brazil
suggests that there is a certain demand there for this type
of firearm. The reported black market price is significantly
higher than in other countries, and also significantly
higher than the price of other firearms in Brazil: around
US$15,000 for an assault rifle, and around $12,500 for
5.56mm calibre part kits. However, these prices remained
only moderately higher than the corresponding licit prices
in Brazil. Such weapons and related part kits were only
licitly available in Brazil to a restricted group of authorized
individuals – shooters, hunters and collectors.
2 Agencia de Informacion Paraguaya, Ministerio de Tecnologias de la
Informacion y Comunicación, Dimabel suspende importación de armas
y municiones para controlar el mercado interno, 23 May 2018 (avail-
able at: https://www.ip.gov.py/ip/dimabel-suspende-importacion-de-
armas-y-municiones-para-controlar-el-mercado-interno/).
FIG. 2
Price of firearms in the illicit market reported by countries in the Americas, 2016-17
* Converted from original currency using UN exchange rates.
Source: UNODC IAFQ.
15,000
0
2,500
5,000
Brazil
Dominican Republic
El Salvador
Peru
Saint Lucia
Brazil
El Salvador
Saint Lucia
Brazil
Costa Rica
Saint Lucia
El Salvador
Peru
Saint Lucia
Saint Lucia
Peru
Pistols Revolvers Rifles Shotguns Submachine
guns
Machine
guns
Price (USD per piece*)
Comparing prices in illicit markets across countries, it
appears that revolvers are less expensive than pistols in the
three countries in Latin America and the Caribbean where
the prices of both arms were available.
In general, firearms prices were relatively low in the illicit
market in El Salvador and Peru.
In Africa, prices in the illicit market were relatively low in
Kenya, including prices of high-powered weapons such as
battle rifles and assault rifles. In Libya, the black market
price of an assault rifle was uncharacteristically low in com-
parison with other types of firearms.
In general, the price data suggest that pistols are frequently
available in the illicit market in a wide range of forms, and
that, while criminals are willing to resort to converted
firearms, they still have a preference for industrially manu-
factured firearms, with niche demand for different kinds
of industrially manufactured pistols. This pattern was most
pronounced in Europe. For example, a converted pistol
in Spain was about six times cheaper than a reactivated
pistol, suggesting lower demand for converted as compared
to reactivated pistols. In Sweden, a converted starting/gas
pistol was reported to cost approximately US$870-1,750,
3
as opposed to $1,170-2,340
4
for an original pistol. In Den-
mark, three different prices were reported for pistols, with
3 7,500-15,000 SEK converted into US$ dollars using average UN
official exchange rates for 2016-17.
4 10,000-20,000 SEK, converted into US$ dollars using average UN
official exchange rates for 2016-17.
Reported value
Midpoint of reported range
Reported range
60
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 4
Price of firearms in the illicit market
reported by countries in Africa, 2016-
17
* Converted from original currency using UN exchange rates.
Source: UNODC IAFQ.
FIG. 3
Price of firearms in the illicit market reported by countries in Europe, 2016-17
* Converted from original currency using UN exchange rates.
** Prices for Croatia represent a lower bound.
Model A and Model B refer to two distinct models from the same manufacturer.
Source: UNODC IAFQ.
used pistols being the least expensive and a specific, smaller
model of pistol – which is easier to conceal and carry – the
most expensive.
With regard to rifles, a notable difference could be seen
between the prices of rifles (including assault rifles) in
illicit markets in the Western Balkans and in Northern
Europe, pointing to the potential of the Western Balkans
as a source region for these firearms. Notably, the price of
a rifle in the illicit Slovak market was lower than the price
for a pistol in the same country.
Beyond Europe, the wide range of prices for pistols is also
conspicuous in the case of Australia, which reported a
range of approximately US$230-6,330
5
for pistols on the
illicit market, encompassing the illicit prices for all the
other types of firearms (also reported as ranges), with the
single exception of high prices of assault rifles reported
from one jurisdiction in Australia.
Modalities of trafficking
Traffickers may resort to different strategies and techniques
to transport firearms across borders or, domestically, from
one location to another. Clearly many factors can influ-
ence these choices, such as the potential for detection, the
ease of concealment, the degree of control exercised at
borders, the feasibility of larger payloads and geographical
convenience.
5 300-8,400 $A, converted into US$ using average UN official exchange
rates for 2016-17.
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
Spain, converted
ƌŽĂƟĂΎΎ
North Macedonia,
ĐŽŶǀĞƌƚĞĚŇĂƌĞŐƵŶ
Denmark,ƵƐĞĚ͕
ŵŝƐĐĞůůĂŶĞŽƵƐƚLJƉĞƐ
Sweden, converted
ŐĂƐŐƵŶͬƐƚĂƌƚĞƌƉŝƐƚŽů
Slovakia
Denmark,DŽĚĞů
Spain
Sweden
Denmark,DŽĚĞů
Lithuania
North Macedonia,
ĂƐƐĂƵůƚƌŝŇĞ
Slovenia,ĂƐƐĂƵůƚƌŝŇĞ
ƌŽĂƟĂΎΎ
Slovakia,ĂƐƐĂƵůƚƌŝŇĞ
Slovenia,
ƐĞŵŝͲĂƵƚŽŵĂƟĐ
Sweden
Denmark,ĂƐƐĂƵůƚƌŝŇĞ
Spain
Slovenia
Denmark
Sweden
PISTOL RIFLE
^hD,/E
'hE
WƌŝĐĞ;h^ƉĞƌƉŝĞĐĞΎͿ
Reported value
Midpoint of
reported range
Reported value
;ĂƐƐĂƵůƚƌŝŇĞͿ
Midpoint of
reported range
;ĂƐƐĂƵůƚƌŝŇĞͿ
Reported range
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
Pistol
Hunting weapon
Pump action gun
Submachine gun
Pistol
Assault rifle, model A
Submachine gun
Battle rifle
Assault rifle, model B
Assault rifle
Shotgun
Pistol
Shotgun
Algeria Kenya Libya Tunisia
Price (USD per piece*)
Reported value
Midpoint of reported range
Reported range
61
Modalities of trafficking 3
FIG. 6
Average distribution* of arms sei-
zures by type of location where they
were seized, 2016-17
* Simple average of distributions for 51 countries, adjusted for seizures
whose type of location was unknown or unclassified.
Source: UNODC IAFQ.
FIG. 5
Number of arms seized in significant
individual cases (common ranges¹),
trafficking cases in comparison with
other cases, 2016-17
1
Interquartile range and full range, excluding outliers.
* A case is considered “small” if 5 arms or less were seized, “large” oth-
erwise. This distinction is made in view of the fact that the threshold of
more than 5 arms seized in a case was one of the criteria recommended
in the questionnaire for the designation of a seizure case as “significant”.
Note: Responses did not always specify whether the cases were traffick-
ing cases or not. Such cases are not included.
Source: UNODC IAFQ.
Planes and
airports, 2.3%
Vessels and
harbours, 1.8%
Land borders, 1.5%
Domestic / within
national territory,
90.8%
Other, 3.6%
Seizures data can provide insights on these trafficking strat-
egies, although it is important to bear in mind that not all
seizures may be associated with trafficking or even with
illicitly sourced firearms. Data on significant seizures
reported on a case-by-case basis show that cases of traf-
ficking tend to involve, on average, larger numbers of fire-
arms per case than cases of firearms that may be stopped
by the authorities for possession or other alleged offences.
In general, the typical (median) number of arms per case
was 10 for trafficking cases and 7 for non-trafficking cases.
The difference is clearer when considering small (less than
five firearms seized) and large (more than five firearms
seized). For small seizures, the median of trafficking cases
is 2 and for non-trafficking cases is 1. For large seizures
the difference is more pronounced: 18 arms is the median
for trafficking cases and 8 for non-trafficking cases (see
Figure 59). However, the fact that the larger number of
arms are seized may impact the likelihood that the seizure
is designated as a case of trafficking.
Cross-border transportation (air/land/
sea/mail) and modi operandi
Data reported on the type of location where arms were
seized indicate that, on average, the overwhelming major-
ity of arms are seized within the national territory (exclud-
ing ports of entry). Cross-border seizures account, on
average, for slightly less than 10 per cent, with the standard
modes of air, sea and land transportation across borders
typically each accounting for around 2 per cent of seizures.
While some of these arms may have been trafficked into
the country and only detected after having crossed the
border, and some diverted domestically (as mentioned
above) some seizures may not always be linked to traf-
ficked, diverted or illicitly sourced firearms and, particu-
larly for small seizures, they may relate to other offences,
such as possession, which may be of an administrative
nature.
Despite the fact that cross-border seizures are a minority
of seized firearms, given their nature, they may be particu-
larly revealing about transnational illicit flows. Based on
seizures by customs authorities, there are clear differences
across the types of conveyance in terms of the total number
of arms seized per case, with seizures from vessels being
on average more than five times larger than any other type
of conveyance. At the same time, seizures from vehicles
accounted for more than two thirds of seizure cases, sug-
gesting that large illegal shipments tend to travel by sea
while small shipments may be more common and tend to
travel by land. Assigning resources to targeting vessels in
law enforcement efforts could significantly impact the
number of firearms intercepted.
International firearms trafficking
methods
More information on trafficking modes is available from
the qualitative assessments provided by countries of the
prevalent modes of transportation for trafficking across
Median
Medi an
Common
(interquartile)
range
All trafficking cases
All non-trafficking cases
Small* trafficking cases
Large* trafficking cases
Large* non-trafficking cases
Small* non-trafficking cases
Medi an
62
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
borders. Countries distinguished land, maritime and air
transportation, as well as mail and other forms of
transportation.
In general, there is a clear distinction between the detec-
tion of incoming and outgoing flows, and most weapons
are seized on an incoming, rather than outgoing, route.
This suggests that the level of scrutiny exercised by cus-
toms authorities tends to be higher on incoming
firearms.
Land borders
Trafficking by land was overall the mode of transportation
most highlighted by countries. Libya, for example,
reported that the most common routes for firearms traf-
ficking was across its eastern, western and southern bor-
ders. Vehicles with false bottoms were used in Mexico and
specific types of vehicles for firearms trafficking was
reported by Costa Rica. Concealment and package deliv-
ery were mentioned as the most frequent modes of trans-
portation in Paraguay. From Paraguay, firearms were
trafficked to Brazil across the land border in cars, trucks
and buses. Land border crossings were also used in Croa-
tia, Slovenia and Denmark. Land borders were mainly
crossed for trafficking purposes also in Albania, either by
vehicles or on foot.
Planes and airports
Eight per cent of firearms seizure cases made by customs
authorities involved trafficking by air with on average, 4
firearms per seizure. Out of 24 countries which reported
FIG. 7
Type of conveyance in customs seizures in terms of size of seizure and number of seizure
cases
* Includes 102 cases of other or unknown type of conveyance.
ª Number in brackets denotes number of cases.
Source: World Customs Organization.
Average number of seized firearms/SALWs per
customs seizure, by type of conveyance, 2016-17
Cases of firearms/SALWs** seized by customs
authorities, by type of conveyance, 2016-17
Vehicle
75%
Mail
10%
Air
8%
Vessel
6%
Pedestrian
1%
any cross-border seizures of arms, 16 had made some of
these seizures in planes and airports. Brazil reported the
detection of a large seizure case of 60 assault rifles, seized
from incoming air freight. In Paraguay, all the arms seized
in 2016
6
were seized in connection with unauthorized
importation attempts by air.
Vessels and harbours
A small number of countries mentioned cross-border traf-
ficking by sea, but among these few reports, there were
very large seizures. For example, Tunisia reported a seizure
of 362 arms in the Port of La Goulette. Trafficking by sea
could be done by hiding firearms in the vessels like was
done in Tunisia were firearms were transported inside vehi-
cles through seaports on board ships from the ports of
Marseille (France) and Genoa (Italy). Another technique
to cross the border avoiding the detection of customs was
reported by the Philippines, whereby firearms were jetti-
soned from vessels at prearranged areas some distance from
the shore and subsequently picked up by small boats. In
addition, leakages in legitimate importations of firearms
were used by traffickers, including some involving the use
of private ports and wharves. In some cases, small boats
were used to cross rivers and lakes like in Brazil where such
border crossings reportedly happen across the Parana river
and Itaipu lake. Another way to transport firearms across
borders by sea was to use undeclared or mis-declared fire-
arms, alongside other goods, addressed to fictitious names
and addresses. In addition, parts and components of dis-
mantled firearms could have been included within
imported or exported metal items or parts of machinery.
6 Seizures in Paraguay amounted to 37 firearms in 2016. No data were
available for 2017.
** Cases of unknown type of conveyance are excluded.
19.8
4.4
2.7
1.9
1.8
3.2
0
5
10
15
20
Vessel
(82)
a
Air
(124)
a
Mail
(139)
a
Vehicle
(1102)
a
Pedestrian
(18)
a
Overall*
(1567)
a
63
Modalities of trafficking 3
Border
Border
Border
Border
d
Borde
Border
Border
Border
Border
Border
Border
Land
TRAFFICKING
Small quantities
in many different
vehicles
Ant
TRAFFICKING
Many actors/traffickers
carrying small amounts
Sea
TRAFFICKING
Big quantities are
smuggled by boats
CROSS-BORDER FIREARMS TRAFFICKING METHODS
FIG. 8
Average distributions* of customs
seizures, by type of routing, 2016-17
* Simple average of data for 59 countries.
Source: World Customs Organization.
Domestic,
4%
Incoming,
71%
Outgoing,
15%
Transit,
10%
Mail and other methods
Traffickers also use methods in which their physical
involvement is reduced and which may be seen as less risky.
Mail was reported to be used for firearms trafficking in
Albania, Slovakia and Lithuania. Traffickers in some coun-
tries used the open internet or the darknet for firearms
trafficking. This included Slovakia, Spain, Libya
7
and in
particular Portugal, where illegal transactions originating
in the European Union and conducted over the darknet
or open internet were detected. Also, Lithuanian authori-
ties indicated having received communication arising from
foreign investigations on Lithuanian nationals suspected
of acquiring firearms illegally on the internet.
Ant trafficking
Cross-border trafficking of firearms sometimes takes on
the form of so-called “ant trafficking”; in other words, the
illicit flow is facilitated by carrying or transporting very
small quantities, possibly by numerous individuals. This
phenomenon has been hypothesized and documented in
various settings and regions, including Africa, Asia,
Europe,
8
and the border between the United States and
7 Some research suggests that the use of online sites and services for
trade in arms in Libya has its beginnings between 2011 and 2013.
See Small Arms Survey, The Online Trade of Light Weapons in Libya.
Dispatch No. 6, April 2016.
8 Savona Ernesto U. and Mancuso Marina (Eds.) 2017. Fighting lllicit
Firearms Trafficking Routes and Actors at European Level. Final Report
By number of cases
of seizures of weapons
and related items
By number of
firearms seized
Domestic,
6%
Incoming,
70%
Outgoing,
13%
Transit,
12%
64
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Border
Border
Ant
TRAFFICKING
Many actors/traffickers
carrying small amounts
ANT TRAFFICKING
Mexico.
9
Such a mechanism may or may not occur in a
concerted fashion or be orchestrated by the final recipients
of trafficked arms, but can also simply be a consequence
of multiple individuals acting independently.
In some cases, cross-border trafficking in small quantities
can be opportunistic and targeted to serve local or small-
scale demand. For example, in its 2013 Serious and Organ-
ized Crime Threat Assessment,
10
Europol assessed that
trafficking in the European context occurred on a small
scale and trafficked weapons were intended for personal
use or to meet specific orders. A joint research initiative
of Small Arms Survey and the African Union Commis-
sion
11
has documented smaller-scale trading of arms, facili-
tated by ethnic ties, by pastoralists, across the borders of
Kenya, Uganda and South Sudan for the purposes of pro-
tection of their herds. This was also corroborated by Kenya
in its reply to the Illicit Arms Flows Questionnaire for
2016-17, which mentioned cattle raids as a context for
arms seizures. In Asia, one instance of cross-border traf-
ficking which has been described as “ant trade” is the cross-
border trafficking into eastern Nepal.
12
Although “ant trafficking” occurs in small individual con-
signments, it can result in sizeable illicit flows and accu-
mulation of illicit firearms, and may also be the result of
an organized strategy or the outcome of a supply and
demand mechanism functioning at a larger scale. One
technique which may be used for this is the use of “mules”,
including migrants and refugees.
13
In its reply to the Illicit
Arms Flows Questionnaire for 2016-17, Libya reported
that arms trafficking routes were often the same as those
used for irregular migration.
of Project FIRE, p. 54, 2017.
9 See, for example: UNODC, The Globalization of Crime – A Transna-
tional Organized Crime Threat Assessment, 2010.
10
Europol, EU Serious and Organized Crime Threat Assessment 2013.
11
Small Arms Survey, Weapons Compass, Mapping Illicit Small Arms
Flows in Africa, January 2019.
12
Small Arms Survey, The Highway Routes. Small Arms Smuggling in
Eastern Nepal. Issue Brief No. 4, November 2014.
13
Small Arms Survey, Weapons Compass, Mapping Illicit Small Arms
Flows in Africa, January 2019.
Another systematic technique to illicitly procure firearms
is the use of straw purchases. This involves an individual,
usually with a clean criminal record, legally buying a fire-
arm with the intention of illegally passing it on to a person
who would otherwise be precluded from owning a firearm,
or whose profile would raise suspicion if they were to
attempt such a purchase themselves. Often the serial num-
bers would be erased and the firearm reported as stolen.
It has been claimed that fake “stealing” of firearms was an
important channel used by organized crime groups in Italy
to secure “clean” weapons for homicides and other crimes.
14
Considering that firearms are available on the licit market,
which is regulated to various degrees, it is often more effec-
tive for criminals to acquire and subsequently divert fire-
arms from the licit market as an alternative to large-scale
trafficking of illicitly sourced firearms. The need to be
discreet – not attracting too much attention from law
enforcement - and to operate (at least in appearance)
within the rules would be compatible with the small num-
bers of weapons procured and moved in single instances,
and would also confirm that the initial procurement may
need to be done by individuals without a criminal record.
Such a mechanism may function both within a given
country and across borders.
The land border between the United States and Mexico
likely represents the earliest instance where the “ant trade”
was observed by scholars.
15
One of the aspects which sug-
gest that this trade is done in an organized fashion is that,
despite the fact that the firearms are moved in smaller
batches, the sources may be more concentrated.
16
Moreo-
ver, a single straw purchase may involve larger quantities
of firearms, and the same individuals may be involved in
multiple straw purchases and multiple border crossings.
17
14 Massari, M. (2013) “Guns in the family. Mafia violence in Italy”, in
LeBrun E., McDonald G., Alvazzi del Frate A., Berman E.G., and K.
Krause, Small Arms Survey 2013, Cambridge University Press.
15
Lumpe, L., ‘The US Arms Both Sides of Mexico’s Drug War’, Covert
Action Quarterly, 61: 39–46, 1998.
16
Cook, P.hilip J., Cukier W. and Krause, K., ‘The illicit firearms trade
in North America’, Criminology and Criminal Justice, 2009; 9; 265-
286.
17
Lumpe, L., ‘The US Arms Both Sides of Mexico’s Drug War’, Covert
Action Quarterly, 61: 39–46, 1998.
65
Modalities of trafficking 3
Using cross-border seizures as a proxy for trafficking, one
indicator which may give further insight into the phenom-
enon of ant trafficking is the extent to which weapons are
seized in smaller quantities, bearing in mind that this may
also be influenced by law enforcement strategies.
The question may arise whether the “ant trade” is the pre-
dominant form of international trafficking. The evidence
from customs seizures does not clearly support the hypoth-
esis of that international firearms trafficking is generally
an “ant” trade. Customs seizures recorded in the World
Customs Organization’s CEN database in 2016-17 show
that firearms trafficking across borders happened at all
levels, including trafficking of individual firearms, a few
firearms and large shipments.
Approximately three quarters of all cases involved only
one firearm, and less than 6 per cent involved more than
5 firearms at a time. However, in terms of numbers of
firearms seized, the top 2 per cent of cases (cases of at least
18 firearms or more) accounted for approximately one half
of all seized firearms. These included 4 cases in which more
than a hundred firearms were seized (see Figure 9). It
should be borne in mind that the small seizure cases may
include many instances which were unrelated to traffick-
ing. On the other hand, the sporadic nature of the very
large instances, in addition to the lack of background
information, makes it hard to reliably interpret their over-
all importance in the big picture of trafficking. While the
characterization of trafficking as an “ant” trade depends
on a subjective threshold of what constitutes a “small”
shipment, this evidence does not warrant describing inter-
FIG. 9
Breakdowns of customs seizures, by size* of seizure case
* The size of a seizure case refers to the number of firearms seized in that particular instance.
a
A case is considered “small” if between 2 and 5 firearms were seized in that particular instance.
b
A case is considered “medium” if between 6 and 10 firearms were seized in that particular instance.
c
A case is considered “large” if between 11 and 17 firearms were seized in that particular instance.
d
A case is considered “exceptionally large” if at least 18 firearms were seized in that particular instance.
Source: World Customs Organization.
Single
firearm,
74%
Small
cases
a
,
20%
Medium
cases
b
, 2%
Large
cases
c
, 1%
Exceponally
large cases
d
, 2%
Breakdown of seizure cases, by size*
national firearms trafficking in general as predominantly
an “ant” trade.
However, a focus on seizures made specifically by customs
authorities in the United States and in Mexico brings out
a striking distinction between such seizures and customs
seizures in general, and confirms that seizures across this
border tend to be significantly smaller than usual. In par-
ticular, cases of 5 firearms or less accounted for two thirds
of firearms seized by Mexico as they crossed the border
from the United States, and an even higher proportion
(78 per cent) of firearms seized by the United States at the
same border (on their way to Mexico). This is in sharp
contrast with customs seizures in other countries, among
which only a quarter of firearms were seized in cases of 5
firearms or less. Moreover, the similarity of the patterns
(see Figure 10) which emerge independently from seizures
by authorities in Mexico and the United States further
corroborates the assumption that these patterns reflect a
real characteristic of flows across this border.
Thus, the comparison between customs seizures in general
and those made specifically en route from the United
States and Mexico does support the hypothesis of ant traf-
ficking between these two countries, in the sense that this
flow appears to occur in smaller individual batches than
the general global pattern.
A comparison of the types of arms seized in the United
States and Mexico also reveals an interesting pattern, with
the share of rifles rising progressively from those recoveries
made in the United States in general (14 per cent), to
Single
firearm,
24%
Small
cases
a
,
17%
Medium
cases
b
, 6%
Large cases
c
,
4%
Exceponally
large cases
d
,
49%
Breakdown of firearms seized,
by size* of seizure case
66
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 10
Size* of customs seizures from the United States to Mexico in comparison with customs
seizures made by all other countries
*The size of a seizure case refers to the number of firearms seized in that particular instance.
Sources: Mexican Customs (data for Mexico); World Customs Organization (data for the United States and all other countries).
FIG. 11
Comparison of the distributions of arms or other weapons seized along the route from
the United States to Mexico, 2016-17
1
Includes, under the category “Other weapons”, the following categories as classified by the Bureau of Alcohol, Tobacco, Firearms and Explosives:
tear gas launchers, destructive devices, combination guns, flare guns and “any other weapon”. Derringers are included under “pistols”. Silencers and
receivers/frames are not included.
2
Includes firearms submitted for tracing to the Bureau of Alcohol, Tobacco, Firearms and Explosives by a law enforcement agency, resulting from
seizure as well as abandonment, buy-back program, or other recovery method. Moreover, only firearms submitted for tracing are included. Firearms
submitted for tracing after recovery do not represent the entire set of all seized firearms.
Pistols
62%
Revolvers
15%
Rifles
14%
Shotguns
9%
Machineguns and sub-machineguns
0.4%
Other weapons
0.1%
Pistols
63%
Revolvers
7%
Rifles
24%
Shotguns
6%
Machineguns and sub-machineguns
0.2%
Other weapons
0.3%
Distribution of firearms seized, by size*
of seizure case
Cumulative proportion of seized firearms,
by size* of seizure case
1 firearm
1 firearm
6 firearms or
more
6 firearms
or more
6 firearms or
more
0% 20% 40% 60% 80% 100%
All other countries
United States,
outgoing to Mexico
Mexico, incoming
from United States
1 firearm 2 firearms 3 firearms 4 firearms 5 firearms 6 firearms
or more
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 4 16 64 256 1,024
Cumulative proportion of seized firearms
Size of seizure case (number of firearms, logarithmic scale)
United States, outgoing to Mexico
Mexico, incoming from United States
All other countries
Arms¹ recovered and traced² by
authorities in the United States
Weapons seized by United States
customs authorities from consign-
ments departing to Mexico Arms seized in Mexico
Source: World Customs Organization. Source: UNODC IAFQ (Mexico).Source: Bureau of Alcohol, Tobacco, Firearms
and Explosives (ATF), United States.
Pistols
39%
Revolvers
7%
Rifles
41%
Shotguns
5%
Machineguns and sub-
machineguns
1%
Other weapons
7%
67
Modalities of trafficking 3
cross-border customs seizures made by authorities in the
United States (24 per cent) and again to seizures in Mexico
(41 per cent). This increasing share is mainly offset by the
smaller share of pistols, but the progression in terms of the
share of pistols is less clear. One possible interpretation for
this pattern is that the illicit market in Mexico exhibits a
strong demand specifically for rifles which may be fed by
flows from the United States. Another interpretation is an
uneven priority given to rifles between the United States
and Mexico at the border.
Domestic trafficking
In general terms, firearms trafficking is understood as the
unauthorized trade, purchase and transportation of fire-
arms. As all other forms of trafficking, firearm trafficking
can happen within and across borders, so trafficking can
be analysed and researched as a domestic and a transna-
tional phenomenon. From a legal perspective, according
to the United Nations Firearms Protocol, “illicit traffick-
ing” of firearms entails a physical movement from the ter-
ritory of one state to that of another state;
18
some countries,
as well as international instruments, refer to broader con-
cepts such as “illicit proliferation”
19
, “diversion” and “illicit
transfers” or “illicit arms flows”(the SDG 16.4), with a
18 See Box ‘International and national definitions of firearm trafficking’.
19
Programme of Action to prevent, combat and eradicate the illicit trade
in small arms and light weapons
MAP 1
Number of arms seized in Central American countries, at sub-national level, 2017
Source: UNODC IAFQ.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
MEXICO
HONDURAS
GUATEMALA
EL SALVADOR
PANAMA
Cayman Islands (UK)
UNITED STATES
COLOMBIA
CUBA
NICARAGUA
COSTA RICA
BELIZE
JAMAICA
HA
ITI
THE BAHAMAS
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
T
otal number of arms seized
0 - 100
101 - 200
201 - 300
301 - 500
>500
No Data
MEXICO
GUATEMALA
HONDURAS
EL SALVADOR
NICARAGUA
BELIZE
view to capture also the domestic aspect of trafficking of
weapons, regardless of their possible or potential subse-
quent crossing of international borders. Some national
legislations, such as Uruguay, for example, have also explic-
itly introduced separate offences for internal and interna-
tional trafficking.
It is difficult to analyse trafficking of firearms by separating
its national and international dimensions. National legisla-
tion may not differentiate trafficking that remains within
national borders and trafficking that comes from abroad
or is destined for abroad, so information related to traffick-
ing, such as seizure data, cannot be disaggregated to reflect
these two aspects separately. Looking at trafficking that
crosses borders in isolation from domestic trafficking would
be misleading because domestic trafficking is often the
beginning of the illicit transnational supply chain of fire-
arms. Hence starting with addressing domestic trafficking
is a way to counter transnational illicit arms flows.
In order to shed light on the illicit movement of firearms
within the territory of a given country, it is useful to exam-
ine seizures at sub-national level (region, state, department,
province, et cetera). A certain level of detail was available
for some countries in Central America.
68
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
MAP 2
Rate of arms seizures per population in Guatemala, Honduras and El Salvador,
in comparison with homicide rates
Source: UNODC IAFQ; UNODC Homicide Statistics.
The boundaries and names shown and the designations used on these maps do not imply official endorsement or acceptance by the United Nations.
EL SALVADOR
HONDURAS
GUATEMALA
MEXICO
NICARAGUA
BELIZE
Arms seized per 100,000 population
30.0
30.1 - 35.0
35.1 - 55.0
55.1 - 75.0
> 75.0
No data
EL SALVADOR
HONDURAS
GUATEMALA
MEXICO
NICARAGUA
BELIZE
omi ides per 100,000 population
10.0
10.1 - 30.0
30.1 - 50.0
50.1 - 70.0
> 70.0
No data
Rate of arms seizures per 100,000 population, 2017
Homicide rates per 100,000 population, 2017/16
69
Modalities of trafficking 3
FIG. 12
Homicide rates and arms seizures per capita
in administrative regions* of Mexico, 2016/17
(ranked correlation)
* 31 states and 1 federal district.
** Data for 2017 were not available.
Sources: UNODC IAFQ; UNODC Homicide Statistics.
Colima
Guerrero
Chihuahua
Sinaloa
Zacatecas
Baja California
Michoacán
Sonora
Morelos
Baja California
Sur
Nayarit
Tabasco
Durango
Tamaulipas
Oaxaca
Guanajuato
Veracruz
Jalisco
San Luis Potosi
Quintana Roo
Puebla
Coahuila de
Zaragoza
Campeche
Hidalgo
Queretaro
Nuevo León
Mexico/Estado de Mexico
Tlaxcala
Aguascalientes
Chiapas
Ciudad de México
Yucatan
R² = 0.60
0
5
10
15
20
25
30
35
0 5 10 15 20 25 30 35
Homicide rate 2016,** ranked (1=lowest, 32=highest)
Arms seizures per capita, 2017, ranked
(1= lowest, 32=highest)
Within the neighbouring countries of El Salvador, Gua-
temala and Honduras, seized arms, as measured in absolute
numbers, tended to concentrate around some of the most
violent areas suggesting that seizures may be broadly linked
with the demand for firearms generated by violent crime.
Data from Guatemala indicate that, in 2017, 80 per cent
of seized arms
20
were seized in the suspected context of
violent crime. For El Salvador, the criminal context was
more frequently assessed to be related to trafficking as
opposed to violent crime
21
, but El Salvador also indicated
that the data pertaining to illicit trafficking included trans-
fer or movement of items within national borders, as well
as cross-border trafficking. Hence these figures do not
allow to distinguish between international movements of
firearms and the final stages of the domestic illicit supply
chain (close to the final “consumers”).
Some of the largest significant seizure cases reported by
these countries also occurred in urban centres. El Salvador
reported a seizure of 110 handguns in Santa Ana. Of the
8 significant cases reported by Guatemala, 5 occurred in
Guatemala City or Guatemala Department, 2 in a port
and 1 on a highway – the largest being a seizure of 29
firearms in Guatemala Department. Of the 7 largest cases
reported by Honduras, 5 were made in San Pedro Sula or
Tegucigalpa, Distrito Central (the capital); this included
a seizure of 68 arms in San Pedro Sula. The largest case,
involving 91 arms, occurred in the municipality of El Pro-
greso – a smaller city but an important logistical hub. The
concentration of seizures in urban areas may suggest a
combination of factors: higher demand of firearms related
to violent crime, concentration of trafficking hubs, and
higher presence and capacity of law enforcement.
When expressed in rates per capita, the most pronounced
seizure levels shift perceptibly from the more populated
areas towards the land and sea borders, with higher rates
registered for example in the departments of La Unión (El
Salvador) and Colón, Copan and Olancho (Honduras).
Border areas are also affected by the highest level of homi-
cide. The presence of trade junctions or logistical hubs
may also play a role, as in the case of Yoro department in
Honduras (including the municipality of El Progreso) and
the two departments, on the Pacific coast of Guatemala,
of Escuintla (with its major port of Puerto Quetzal) and
Santa Rosa (also close to the land border and including
the municipality of Chiquimulilla). Thus, the seizure levels
expressed per capita are more likely than the absolute
values to reflect the transnational movement of firearms
among these countries as well as neighbouring countries.
El Salvador reported sending and receiving large numbers
of tracing requests from neighbouring countries.
20 3,651 out of a total of 4,686.
21 Out of 3,103 weapons seized, only 345 were suspected to be linked
to a context of violent crime, compared to 1,898 weapons linked to a
context of trafficking.
In Mexico, among the 32 administrative regions (31 states
and one federal district), the highest numbers of arms
seized in 2017 were seized in Tamaulipas, Sinaloa, Micho-
acán and Guerrero. All four states were also among the
top five (along with Baja California Sur) when ranked in
per capita terms. In general, there appeared to be an asso-
ciation between homicide rates and seizures of arms per
capita (see Figure 12). Overall, this suggests a link to the
presence of organized crime groups and inter-cartel vio-
lence. There are some exceptions to this general pattern as
in the case of Tamaulipas, which was the most prominent
in both absolute and per capita seizure terms, but with
comparatively moderate levels of homicidal violence; the
proximity to the border with the United States may bring
about a significant role in firearm trafficking. Tamaulipas
and Guerrero were also the location of large significant
seizures reported by Mexico: 101 rifles seized in Tamauli-
pas, and 48 arms seized in Guerrero (the only other sig-
nificant case reported by Mexico was, however, even larger,
involving 109 arms seized in Nuevo León).
Some detailed data at sub-national level were also available
for the United States, in particular information on the
states to which recovered weapons could be traced. These
data enable a quantification of the extent to which recov-
ered weapons had been sold in a specific state, both among
that state’s own recoveries as well as recoveries in other
states. These two independent measures result in similar
outcomes in terms of the importance of states relative to
70
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
MAP 3
States in the contiguous United States,¹ as sources for domestically traced firearms,² 2017
1
District of Columbia is also included. Alaska, Hawaii, Guam, Puerto Rico and the United States Virgin Islands are not included as they do not share a
land border with other parts of the territory of the United States.
2
May include tear gas launchers, destructive devices, receivers/frames, silencers and any other weapons classified as “firearms” by the Bureau of Alcohol,
Tobacco, Firearms and Explosives (ATF), United States.
Note: Purchases refer to the legal market. Based on a subset of traced firearms for which it was possible to identify a purchaser and the state in which
the final dealer was located. Firearms submitted for tracing to the Bureau of Alcohol, Tobacco, Firearms and Explosives by a law enforcement agency
may result from a seizure as well as abandonment, buy-back program, or other recovery method. Firearms submitted for tracing after recovery do not
represent the entire set of all seized firearms.
Source: Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF), United States.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
CANADA
MEXICO
CUBA
roportion of re o ered
firearms i ere
pur ased in same state
60
60.1 - 70
70.1 - 0
> 0
umber of firearms
pur ased in t e state and
re o ered in ot er states
2,000
(Si e of the circle is proportional to the number of firearms)
each other. In other words, those states which emerge as
the most prominent on the basis of their own recoveries
– in the sense that they recovered relatively high propor-
tions of firearms originally sold within the same state –
tended to be the same ones which emerged most
prominently as the location of purchase of firearms recov-
ered in other states. The fact that these independent metrics
by and large tally strengthens their interpretation as a meas-
ure of the varying degrees to which firearms involved in
crime can be sourced from a given state. States with a higher
proportion of firearms recovered within the state and high
number of firearms purchased in the same states are more
likely to be the source of domestic trafficking (see Map 3).
Overall, among firearms for which it was possible to iden-
tify a purchaser and the state in which the final dealer was
located, around 71 per cent had been purchased in the
same state where they were recovered.
Information on the time elapsed in the United States
between the first retail purchase of a recovered and traced
firearm and the moment it is recovered indicates that sev-
eral years may pass from the time when a firearm is legally
purchased until it is seized, depending on the geographical
distance between the States where the firearm is purchased
and intercepted. Data indicate that, among traced firearms
for which it was possible to identify the first retail purchase
(including the date of purchase), around 60 per cent had
been sold more than three years before being recovered.
Moreover, the proportion of firearms which were recovered
in a state other than that in which they had been sold
increased steadily with the time elapsed between sale and
recovery. These data suggest that, not only can firearms
last a long time in circulation before they are used and
detected in a crime, but also that the longer they are in
circulation, the more they are likely to move geographi-
cally, at least within the same country.
Transnational trafficking flows
Respondents to the Illicit Arms Flows Questionnaire pro-
vided information about the routes that were observed in
seizures related to illicit trafficking of firearms. Informa-
tion about the countries of departure and intended desti-
nation of seized consignments was also included in the
Customs Enforcement Network database maintained by
71
Modalities of trafficking 3
FIG. 13
Firearms* recovered in the United States and traced domestically to first retail purchase, 2017
* May include tear gas launchers, destructive devices, receivers/frames,
silencers and any other weapons classified as “firearms” by the Bureau of
Alcohol, Tobacco, Firearms and Explosives (ATF), United States.
Note: Purchases refer to the legal market. Based on a subset of traced firearms for which it was possible to identify a purchaser and the state in which
the final dealer was located, and to determine the time elapsed between purchase and recovery. Firearms submitted for tracing to the Bureau of Alco-
hol, Tobacco, Firearms and Explosives by a law enforcement agency may result from a seizure as well as abandonment, buy-back program, or other
recovery method. Firearms submitted for tracing after recovery do not represent the entire set of all seized firearms.
Source: Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF), United States.
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
Under
3
months
3-6
months
7
months
-
under
1 year
1 year -
under
2 years
2 years
-
under
3 years
3 years
and
over
Number of firearms
Time elapsed between purchase and recovery
0%
20%
40%
60%
80%
100%
Under 3
months
3-6
months
7
months -
under 1
year
1 year -
under 2
years
2 years -
under 3
years
3 years
and
over
Proportion purchased and recovered
in same state/neighbouring states/non-neighbouring
states (percentage)
Time elapsed between purchase and recovery
Purchased and recovered in non-neighbouring states/territories
Purchased and recovered in neighbouring states/territories
Purchased and recovered in same state/territory
the World Customs Organization. Although these data
were of a different nature, their combined coverage could
be used to build a picture of illicit trafficking flows. In
general, the information about incoming seizures and
flows was more comprehensive than information about
outgoing seizures.
Northern America, Europe and Western Asia emerged
prominently as points of departure for illicit flows – in
different ways. Northern America was important both in
terms of trafficking to other parts of the Americas, notably
Central America, as well as inter-regional flows. The role
of Europe as a source was more pronounced in terms of
inter-regional flows. Considering inter-regional flows only,
illicit flows departing from Northern America and Europe
together were estimated to account for the vast majority
of the global total, with Western Asia the only other sub-
region accounting for a non-negligible share (see Map 5
and corresponding figure in the Methodological Annex).
Western Asia (along with South America) was important
mainly in terms of trafficking within the subregion, but
also stood out in terms of the diversity of the destination
regions where it was identified as a source for illicit flows.
The same characteristic was even more marked for North-
ern and Western Europe and was also striking—despite
the overwhelming intra-regional component—for North
-
ern America.
The most important areas of destination were Central
America and, taking into account trafficking within sub-
regions, Western Asia and South America. These main
destination areas are known for high levels of criminal
violence or conflict.
Overall, illicit flows appeared to have a certain localized
character. For most regions, the inter-regional component
of incoming flows (that is, illicit flows originating within
the same region) accounted for the large majority of
incoming illicit flows, with the exception of Eastern and
South-East Asia, Oceania and Western and Middle Afri-
ca.
22, 23
In some subregions–namely Northern and Western
Europe, South America and Western Asia–the localized
character was even more pronounced, in that a majority
of incoming flows were detected on their way from coun-
tries within the same subregion.
22 Once more, for these exceptions, the inter-regional flows originated
mainly from the three main source regions of Northern America,
Europe and Western Asia. It should, however, be borne in mind that
coverage from Asia and Africa was relatively limited.
23
It should however be borne in mind that coverage from Asia and
Africa was relatively limited.
Number of firearms,* by time elapsed between purchase
and recovery
Distribution by geographic proximity of purchase
to recovery, according to time elapsed
72
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
MAP 4
Main transnational firearms trafficking flows (as defined by routes of seized firearms),
2016-17
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
Subregions
Eastern Europe
Northern AfricaCentral America
Caribbean
South America
Northern America
Northern and Western Europe Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
sĞƌLJƐŵĂůůǀŽůƵŵĞŇŽǁ
Small
Medium
Large
Southern Europe (excl. WB)
This local character could also be observed in the reports
of specific countries. For example, Sudan reported incom-
ing routes from five different countries, all of which share
a land border with the country. Sudan reported 32 signifi-
cant cases for which the country of departure was believed
to be Libya (the largest three seizures made in Sudan arriv-
ing from Libya involved 234, 86 and 53 firearms). Algeria
also detected and seized firearms which had departed from
bordering countries, including Libya – although countries
outside Africa were assessed to account for larger shares of
incoming flows.
Although the available seizure data do not provide solid
evidence of large illicit flows of arms affecting Africa, this
may also be a reflection of lower capacity to intercept and
record transnational shipments. It should also be borne in
mind that, for some countries, the proliferation of arms
and their ample supply on the black market may poten-
tially render transnational trafficking, especially inter-
regional trafficking, largely unnecessary as a way to illicitly
source firearms.
24
Nevertheless, it appears that Libya may
be an important source country for arms trafficked to
neighbouring countries.
24 See Small Arms Survey, Weapons Compass, Mapping Illicit Small Arms
Flows in Africa, January 2019, p. 38.
Another example of the local character of trafficking is the
case of Brazil, although in this case it is the interplay with
the licit sphere which brings in an inter-regional element.
Brazil identified a number of countries in South America
as the country of departure in seizures related to traffick-
ing, most prominently Paraguay. The routes into Brazil
involved neighbouring countries in various ways, including
diversion from holdings of state authorities, importation
from third countries for the purposes of diversion and traf-
ficking into Brazil, and also trafficking originating in the
United States going through Paraguay as a transit country.
While trafficking of firearms into Brazil appeared to be
mainly done alongside drugs and contraband, in their illicit
shipments, there were also groups mainly dedicated to fire-
arms trafficking. Some groups based in São Paulo and Rio
de Janeiro – the two most important destinations for traf-
ficked firearms within Brazil - also set up bases in Paraguay
to receive drugs from the Andean countries and also
manage the firearms trafficking from Paraguay.
Brazil reported several significant seizure cases of firearms
believed to have entered the country from Paraguay, includ
ing 4 cases of more than 30 firearms each
25
- mainly pistols
- as well as a case of 61 rifles trafficked from the United
States. Moreover, two significant seizures were made in the
73
Modalities of trafficking 3
city of São Paulo in which the seized firearms (9 and 19)
were believed to have entered Brazil from the Plurinational
State of Bolivia and were seized alongside large quantities
of cocaine (882 kg and 273 kg respectively).
26
Brazil also reported on a pattern in trafficking which has
in recent years been successfully countered and largely
reduced through efforts undertaken by the state. The pat-
tern involved firearms being legally exported to neigh-
bouring countries and subsequently trafficked back into
the country; the introduction in 2001 in Brazil of a tax
on exports to some countries in South America appears
to have made it less viable for traffickers to use this strat-
egy, although there may remain instances of “triangula-
tions”, in which firearms are legally exported to a third
country outside South America and subsequently re-enter
Brazil through neighbouring countries. The latter modal-
ity however appears to occur on a small scale and to fall
within a broader pattern of channelling firearms from
licit sources into the Brazilian black market through or
25 One of these cases, involving firearms held in a private home on behalf
of a drug trafficking organization, was not designated as a case of traf-
ficking.
26
These two cases were not designated as trafficking cases.
from neighbouring countries, predominantly from the
legal market of third countries outside the region but also
directly from the stockpiles held by authorities of some
neighbouring countries.
Reports of trafficked firearms departing from the Western
Balkans and detected outside this subregion were limited
and mainly restricted to neighbouring countries (Slovenia
and Croatia), in addition to Sweden. In terms of seizures
made by countries in the Western Balkans themselves,
27
customs seizure data provided additional, but still limited,
evidence of flows departing from the Western Balkans to
other regions (4 cases amounting to 34 firearms in
2016-17).
27 In its reply to the Illicit Arms Flow Questionnaire for 2016-17,
Albania also reported seizures to unspecified destinations.
MAP 5
Inter-regional firearms trafficking flows (as defined by routes of seized firearms), 2016-17
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
Regions
Europe
Americas
Oceania
Asia
Africa
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
74
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
MAP 6
Transnational firearms trafficking flows affecting Latin America and the Caribbean (as
defined by routes of seized firearms), 2016-17
Eastern Europe
Northern Africa
Central America
Caribbean
South America
Northern America
Northern and Western Europe Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
Subregions
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
Southern Europe (excl. WB)
FIG. 14
Estimated breakdown of firearms seized from incoming consignments by subregion of departure,
according to subregion of seizure, 2016-17
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
Subregion of seizure
Western
Balkans
Western
Asia
Western and
Middle Africa
Southern Europe
(excl. WB)
South
America
Oceania
Northern and
Western Europe
Northern
America
Northern
Africa
Eastern
Europe
Eastern and
South-Eastern Asia
Central
America
Caribbean
Caribbean
Central America
Eastern and
South-Eastern Asia
Eastern Europe
Northern Africa
Northern America
Northern and
Western Europe
Oceania
South America
Southern Europe (excl. WB)
Western and Middle Africa
Western Asia
Western Balkans
Rest of the world
Central America
Central America
Central America
Eastern and
South-Eastern Asia
Eastern Europe
Eastern Europe
Northern America
Northern America
Northern America
Northern America
Northern America
Northern America
Northern America
Northern America
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
Northern and
Western Europe
South America
Western Asia
Western Asia
Western Asia
Western Asia
Western Asia
Western Asia
Western Asia
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
0%
%
20% 40% 60% 80% 100%
Subregion of departure
75
Modalities of trafficking 3
MAP 7
Transnational firearms trafficking flows affecting Northern and Central America (as defined
by routes of seized firearms), 2016-17
Subregions
Eastern Europe
Northern AfricaCentral America
Caribbean
South America
Northern America
Northern and Western Europe
Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
Southern Europe (excl. WB)
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on these maps do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
MAP 8
Transnational firearms trafficking flows affecting Europe (as defined by routes of seized
firearms), 2016-17
Subregions
Eastern Europe
Northern Africa
Central America
Caribbean
South America
Northern America
Northern and Western Europe Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
Southern Europe (excl. WB)
76
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Subregions
Eastern Europe
Northern Africa
Central America
Caribbean
South America
Northern America
Northern and Western Europe
Western Balkans
Oceania
Western Asia
Western and Mddle Africa
Rest of the World
Eastern and South Eastern Asia
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
Southern Europe (excl. WB)
MAP 9
Transnational firearms trafficking flows affecting Asia and Oceania (as defined by routes of
seized firearms), 2016-17
The breakdown into subregional groupings is based on the standard UN classification (M49), adapted to take into account the availability of data and
regions of special interest of the study. Please see Methodological Annex for details.
Arrows represent flows between subregions (not specific countries).
Source: UNODC elaboration of data from Illicit Arms Flows Questionnaire and World Customs Organization.
The boundaries and names shown and the designations used on these maps do not imply official endorsement or acceptance by the United Nations.
The final status of Jammu and Kashmir has not yet been agreed upon by the parties. A dispute exists between the Governments of Argentina and the
United Kingdom of Great Britain and Northern Ireland concerning sovereignty over the Falkland Islands (Malvinas).
MAP 10
Transnational firearms trafficking flows affecting Africa and Western Asia (as defined by
routes of seized firearms), 2016-17
Eastern Europe
Northern AfricaCentral America
Caribbean
South America
Northern America
Northern and Western Europe Western Balkans
Oceania
Western Asia
Western and Middle Africa
Rest of the World
Eastern and South Eastern Asia
dŝŶLJǀŽůƵŵĞŇŽǁ
sĞƌLJůĂƌŐĞǀŽůƵŵĞŇŽǁ
Very small
Small
Medium
Large
Southern Europe (excl. WB)
Subregions
77
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
CHAPTER 4
LINKS TO
ORGANIZED CRIME,
TERRORISM AND
OTHER FORMS OF
CRIME
Chapter overview
This chapter takes a step back to analyse the bigger picture
of linkages between firearms trafficking and a range of
other crimes and social issues. Firearms trafficking often
takes place to satisfy demand from criminals who need the
arms for use in various unlawful activities. Drug traffick-
ing and violent crime – including its starkest manifesta-
tion, homicide – are among the key crimes associated with
the circumstances of firearms seizures.
There seems to be a relationship between the level of cer-
tain crimes registered by countries and the share of firearms
that are seized in those criminal contexts. Although there
are some variations, in general, the higher the homicide
rate or rate of illicit drug seizures, the larger the share of
firearms seized in those criminal contexts. In addition,
illicit drugs are the most common non-firearms-related
commodities seized together with firearms, followed by
counterfeit goods, cultural property and natural resources.
In several countries, the use of firearms is particularly pro-
nounced in organized crime or gang-related homicides.
While the majority of homicide victims globally are men,
this trend is typically even more acute when considering
homicides perpetrated by firearms. However, data for a
limited number of countries related to homicides of inti-
mate partners and family members – in which most vic-
tims are women - show other gender-related patterns. Men
were more likely than women to use a firearm when killing
their female partners, while women were more likely to
resort to a sharp object.
More than half (54 per cent) of all homicides in 2017 were
carried out with firearms, and the availability of firearms
seems to have an impact on homicide rates. An increase
in the rate of possession of firearms broadly corresponds
to an increase in the homicide rate. This can best be
observed when comparisons are made within relatively
homogeneous country groupings, for example in terms of
geography or socio-economic characteristics. Comparing
firearms seizures and homicide rates can also help to iden-
tify countries with better firearms interdiction capacity
and stronger rule of law. Countries with higher levels of
firearms seizures relative to the number of homicides com-
mitted with firearms are those where higher seizure levels
may be related to the capacity of law enforcement rather
than firearm supply. These countries tend to have lower
overall levels of homicides, possibly reflecting an overall
strong rule of law situation. Data also show some correla-
tion between firearm ownership and firearm homicide.
Statistical models suggest that a 1 per cent increase in the
rate of firearm civil possession can bring a 1.13 per cent
increase in the homicide rate in the case of developing
countries and 0.74 per cent in the case of developed
countries.
78
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
and seized weapons. Aside from weapons-related items
such as ammunition, parts and components, and explo-
sives, drugs emerge as the most common commodity
seized together with firearms, followed by counterfeit
goods, cultural property and natural resources.
Some countries also provided contextual information
which provide further insight into the circumstances of
weapons seizures and the links to various forms of organ-
ized crime.
Albania reported three significant seizure cases (among
others), over the period 2016-17, that involved the seizure
of large quantities of cannabis in conjunction with several
rifles. North Macedonia reported on the procurement of
weapons, ammunition and explosive devices by individu-
als or groups connected with various criminal activities,
including property crimes, violent crime, illegal hunting
and the illegal drug trade. In particular, North Macedonia
reported a significant seizure of 13 weapons, including 3
rifles as well as a hand grenade and two chemical bombs,
which was linked to an organized crime group connected
to migrant smuggling and drug trafficking. Moreover,
weapons were often found in small numbers in apartment
and house searches related to the detection of crimes such
as drug trafficking, attempted homicide or acts against
public safety. Illegally held firearms were found in the pos-
session of individuals who had committed offences includ-
ing homicide, extortion, grand theft and robbery.
FIG. 1
Average proportion of arms seized in
the suspected context of violent crime,
by rank of homicide rate of reporting
countries, 2016*
a
6 countries with homicide rate of less than 1.05 per 100,000 popula-
tion.
b
6 countries with homicide rate in the range of 1.05-6 per 100,000
population.
c
6 countries with homicide rate of more than 6 per 100,000 population.
* In the case of one country, the homicide rate for 2016 was not avail
-
able, and the corresponding rate for 2015 was used as a proxy.
Note: Reported values of 0 arms seized in the context of violent crime
are not included.
Sources: UNODC IAFQ; UNODC Homicide Statistics (UN-CTS and
other official sources).
In order to adequately explore the links with various forms
of crime, this chapter draws substantially on data sources
which go beyond the Illicit Arms Flow Questionnaire.
This includes other data collection exercise administered
by UNODC—namely the Annual Report Questionnaire
(for data on drug seizures) and the Crime Trends Survey
(for data on homicide)—specific data collection initiatives,
published studies (including the UNODC Global Study
on Homicide) and estimates of civilian holdings of firearms
published by Small Arms Survey.
Forms of crime linked
to firearms
As discussed in Chapter 1, reported data confirm the link
between crime and firearms, since the criminal context of
seized firearms extends well beyond firearms-related offenc-
es.
1
Given that these data are based on recorded suspicions
of non-firearms-related offences, they likely underestimate
the real extent of the incidence of such offences in the con-
text of firearms seizures. Nevertheless, from these data, drug
trafficking and violent crime emerge as the more prominent
types of criminal activity associated with the circumstances
of seized firearms outside of firearm-related offences.
Regional variations can be observed in the different kinds
of criminal context seen in these data. As a context for
arms seizures, violent crime is most pronounced in Africa
and in Latin America and the Caribbean. This is in line
with the relatively high levels of violent deaths (conflict-
related or otherwise) in these regions, including intentional
homicide and specifically firearm-related homicides.
2
Drug
trafficking is also prominent in Latin America and the
Caribbean, where the links between this phenomenon and
violence are also well-documented.
3
Relative to other kinds
of criminal context, drug trafficking is also prominent in
Europe, closely followed by violent crime, while terrorism
is most pronounced in the case of Africa (see Figure 4 in
Chapter 2).
The association between violent crime as a context for
firearms seizures and violence in general can be corrobo-
rated with UNODC’s homicide statistics. Similarly, data
on drug seizures corroborate an association between drug
trafficking as a context for firearms seizures and drug traf
-
ficking in general (see Figures 1 and 2). The countries
especially affected by violent crime and drug trafficking
are reflected in a correspondingly higher share of arms
seized in this context.
Furthermore, data on significant cases of firearms seizures
also corroborate the strong link between drug trafficking
1 See Figure 3 in Chapter 2.
2 UNODC, Global Study on Homicide, 2019.
3 UNODC, Global Study on Homicide, 2019; UNODC, World Drug
Report, 2016.
0%
5%
10%
15%
20%
Lower ª
homicide
rate
Medium
b
homicide
rate
Higher
c
homicide
rate
Proporon (percentage)
79
Forms of crime linked to firearms 4
Denmark reported that firearms were often trafficked in
conjunction with, or in return for, illicit drugs. A number
of firearm-related cases exhibited the involvement of street
gangs or motorcycle groups. On several occasions, authori
-
ties observed the pooling of firearms and ammunition,
stored in neutral locations known only to key gang mem-
bers and used in multiple shootings.
Peru reported on the involvement of cartels in the illegal
importation of firearms by air to be delivered to criminals
in areas affected by illicit coca bush cultivation, such as
the Valle de los Ríos Apurímac, Ene y Mantaro. The
intended recipients included terrorists as well as drug traf-
fickers seeking weapons as a way to enhance their own
“safety” and to enable their confrontations with law
enforcement. Moreover, corrupt officers in the National
Police and Armed Forces diverted weapons from their
holdings. Firearms diverted from the armed forces of other
countries also entered Peruvian territory to be used in vari-
ous forms of crime, including organized crime and drug
trafficking.
Brazil reported on the parallel trafficking of drugs and fire-
arms along two main routes: across the land border from
Paraguay (where marijuana is trafficked along with
cocaine); or on small planes from the Plurinational State
of Bolivia to the countryside of states in the southeast of
Brazil, such as São Paulo or Minas Gerais, from where some
of the illegal goods were further trafficked to Rio de Janeiro.
Ultimately most of the trafficked firearms were intended
for big cities such as São Paulo and Rio de Janeiro. In the
case of Rio de Janeiro, one of the main drivers of demand
for firearms was their use in maintaining control of areas
for drug dealing. The criminal groups who received the
firearms in São Paulo tended to manage the illicit circula-
tion and further distribution of these firearms by renting,
lending and selling them, for the purpose of committing
crime such as bank robberies.
In general, there were people specialized in the transport of
drugs, contraband goods and firearms from the land border
to the urban centres, via the Brazilian highways. Some crim-
inal groups based in São Paulo and Rio de Janeiro involved
in importing drugs from Andean countries had also set up
cells in Paraguay to manage the firearms trafficking from
this country. Although in many cases criminal groups sent
firearms together with drugs in their illicit shipments, there
were also groups dedicated mainly to firearms trafficking.
Libya reported a certain overlap in the routes used for
firearms trafficking and migrant smuggling. These
included routes through several areas on or close to the
borders with Algeria, Niger, Chad, Sudan and Egypt.
Kenyan authorities, in reporting on the context of signifi-
cant cases of firearms seizures, referred to armed robberies,
carjacking, cattle raids, terrorism and inter-community
clashes.
FIG. 2
Average proportion of arms seized
in the suspected context of drug
trafficking, by rank* of drug seizures
per capita of reporting countries, 2016
a
7 lowest-ranked countries (among 20).
b
6 medium-ranked countries (among 20).
c
7 highest-ranked countries (among 20).
* Based on the average of 4 separate rankings for each of the four major
drug classes: amphetamine-type stimulants; cannabis; cocaine-type; and
opiates. Within each drug class, aggregate weights per capita were based
on representative drug types of roughly interchangeable weight, chosen
as follows: amphetamine and methamphetamine for amphetamine-type
stimulants; marijuana and cannabis resin for cannabis; all forms of
cocaine (including base, paste and salts, but excluding coca leaf and coca
bush) for cocaine-type; heroin and morphine for opiates.
Note: Reported values of 0 weapons seized in the context of drug
trafficking are not included.
Sources: UNODC IAFQ; UNODC Annual Report Questionnaire (ARQ)
and other official sources (drug seizures).
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
Lower ratesª
of drug seizures
per capita
Medium rates
b
of drug seizures
per capita
Higher rates
c
of drug seizures
per capita
Proporon (percentage)
FIG. 3
Proportion of significant firearm
seizure cases involving other com-
modities, by type of commodity,
2016-17
Source: UNODC IAFQ.
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Any commodity other than firearms
Ammunition
Other
Parts and components
Explosives
Drugs
Counterfeit goods
Cultural property
Natural resources
Money
Contraband goods
Precursors
Any commodity
other than firearms
80
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Weapons as a nexus between crime and terrorism (cooperation between
criminal and terrorist actors in weapons exchange)
Both criminals and terrorists rely to a large extent on
weapons for their activities; thus the procurement of
firearms is potentially a natural area of cooperation
between the two types of actors.
An initiative of the University of Massachusetts Lowell
documented instances of cooperation between crime
and terror networks (on a global basis), of which 88
instances included cooperation specifically in the area
of weapons exchange. Among these instances, sev-
eral other areas of cooperation were observed, nota-
bly (in more than half the cases) financial support but
also information sharing, joint operations, provision
of training and provision of a safe haven.
With respect to certain specific areas of cooperation
- namely financial support, provision of training, pro-
vision of a safe haven, as well as weapons exchange
itself - the initiative also documented the “direction”
of assistance, that is, whether it was the crime net-
work providing support or assistance to a terror net-
work, or the other way around. From this information
it clearly emerges that terrorist organization rely heav-
ily on criminal organizations to support their opera-
tions, including in the provision of firearms.
In about three quarters of cooperation initiatives, it
was a crime network which provided weapons to a
terror network. Moreover, even among these
instances, other forms of support also tended to
occur in the same direction; for example, there were
about three times as many cases of financial support
from a crime network to a terror network as there
were the other way around, despite the fact that
financial support might be expected to compensate
for the provision of weapons.
The acquisition of firearms by terrorists was also
explored, in the European context, in a study* con-
ducted by the Flemish Peace Institute which drew on
interviews with more than a hundred criminal justice
practitioners from eight European countries. The
study confirms that terrorists’ access to the criminal
milieu is key for the acquisition of firearms, especially
in the case of religiously-inspired terrorist attacks (as
opposed to separatist organizations or extremist
groups with a political inspiration). The study observes
that most perpetrators of recent religiously-inspired
terrorist attacks in the European Union appear to
appear to have been involved in low-level criminality
rather than organized crime, and, in order to procure
their firearms, tended to leverage the connections to
the criminal underworld—along with the familiarity
with firearms—acquired prior to their radicalization.
The criminal suppliers often may not have been aware
of the intentions of the terrorists; indeed, the study
suggests that criminals, especially organized crime
groups, may be averse to supplying terrorist groups
with firearms due to the attention it may draw from
law enforcement authorities.
* Flemish Peace Institute, Firearms acquisition by terrorists in Europe. Research
findings and policy recommendations of Project SAFTE,. Flemish Peace Insti-
tute, Brussels, April 2018.
Instances of additional areas of cooperation between
crime and terror networks, among instances of cooperation
in the provision of weapons
Note: Based on 88 documented instances of collaboration between a crime-
network and a terror network, all of which included collaboration in the
provision of weapons.
Source: University of Massachusetts Lowell, Crime-Terror Collaboration
Database.
Breakdowns of instances of cooperation between crime
and terror networks in the provision of weapons, according
to additional type of support and by direction of provision
Source: University of Massachusetts Lowell, Crime-Terror Collaboration
Database.
Note: Based on 88 documented instances of collaboration between a crime-
network and a terror network, all of which included collaboration in the provi-
sion of weapons.
0%
10%
20%
30%
40%
50%
60%
70%
Financial
Information
sharing
Joint
operations
Training
Safe haven
Development
of technology
Percentage
Additional area of cooperation
ŽŽƉĞƌĂƟŽŶŝŶƚĞƌŵƐŽĨƉƌŽǀŝƐŝŽŶŽĨǁĞĂƉŽŶƐ
Weapons:
Crime to
Terror
Weapons:
Terror to
Crime
Weapons:
Crime to
Terror
Weapons:
Terror to
Crime
35%
14%
9%
2%
7%
2%
5%
11%
28%
11%
59%
16%
WƌŽǀŝƐŝŽŶŽĨǁĞĂƉŽŶƐǀĞƌƐƵƐĮŶĂŶĐŝĂůƐƵƉƉŽƌƚ
WƌŽǀŝƐŝŽŶŽĨǁĞĂƉŽŶƐǀĞƌƐƵƐƚƌĂŝŶŝŶŐ
Weapons:
Crime to
Terror
Weapons:
Terror to
Crime
14%
2%
5%
2%
59%
18%
WƌŽǀŝƐŝŽŶŽĨǁĞĂƉŽŶƐǀĞƌƐƵƐƐĂĨĞŚĂǀĞŶ
ŽŽƉĞƌĂƟŽŶŝŶƚĞƌŵƐŽĨƉƌŽǀŝƐŝŽŶŽĨƐĂĨĞŚĂǀĞŶ
Safe haven:
Crime to Terror
Safe haven:
Terror to Crime
No provision
of safe haven
ŽƚŚĨŽƌŵƐŽĨĐŽŽƉĞƌĂƟŽŶ
ŝŶƐĂŵĞĚŝƌĞĐƟŽŶ
ŽŽƉĞƌĂƟŽŶŝŶŽƉƉŽƐŝƚĞ
ĚŝƌĞĐƚŽŶƐ
ŽŽƉĞƌĂƟŽŶŽŶůLJŝŶƚŚĞ
ƉƌŽǀŝƐŝŽŶŽĨǁĞĂƉŽŶƐ
ŽŽƉĞƌĂƟŽŶŝŶƚĞƌŵƐŽĨĮŶĂŶĐŝĂůƐƵƉƉŽƌƚ
ŽŽƉĞƌĂƟŽŶŝŶƚĞƌŵƐŽĨƉƌŽǀŝƐŝŽŶŽĨƚƌĂŝŶŝŶŐ
Financial support:
Crime to Terror
Financial support:
Terror to Crime
EŽĮŶĂŶĐŝĂů
support
Training:
Crime to Terror
Training:
Terr or t o C rim e
No provision
of training
81
Forms of crime linked to firearms 4
The situational context of a homicide is classified in
UNODC data into the following four main categories:
homicides committed by intimate partners or family mem-
bers of the victim (IPFM homicides, typically occurring
in a domestic or family setting); homicides occurring
during the perpetration of robbery; homicides committed
by members of an organized group or in the context of
organized crime; and homicides related to gang activity.
Although breakdowns of homicide data by situational con-
text are collected by UNODC, the additional layer of
detail required in the data to assess the association between
firearms and the situational context of homicides was not
available on a systematic basis. Nevertheless, data from a
limited number of countries do suggest some recurring
patterns. For example, data from Canada and El Salvador
indicate that the role of firearms was the least marked
among IPFM homicides, while homicides related to organ-
ized crime or gangs were much more likely to involve
firearms.
Some other countries provided parallel breakdowns of
homicide, by situational context and by mechanism
(including firearm), which were sufficiently comprehensive
as to allow to indirectly evince (subject to certain assump-
tions) estimated breakdowns by mechanism within each
Japan reported a total of 8 significant cases of firearms sei-
zures, all of which were made from members of various
Japanese organized crime groups (“Boryokudan”).
Firearms as a mechanism for homicide
Firearms play a prominent role as an instrument used in
the perpetration of homicide. Globally, UNODC esti-
mates that, in 2017, a majority of homicides (54 per cent)
involved firearms, and this proportion is even higher in
certain countries, reaching around three quarters in the
Americas overall. Some countries with high proportions
of firearms-related deaths tend to have high rates of homi-
cide, suggesting that firearms are key enablers of high hom-
icide levels.
Triangulating data from different collection streams, a pat-
tern can be discerned by expressing the number of firearms
seized in relation to the number of homicides committed
by firearms. Higher rates of firearm seizures versus firearms
homicides can be observed in countries with low homicide
rates, while low rates of firearms seizures versus firearms
homicides are observed in countries experiencing high
levels of homicide (see Figure 5). This pattern suggests
that relative higher levels of vigilance exercised in inter-
cepting firearms go along with low homicide levels. It
doesn’t suggest a direct causal relationship between seizures
and homicide but it highlights a possible underlying expla-
nation with the rule of law maintaining a low homicide
rate and ensuring high vigilance on firearms.
Homicides occur in a variety of contexts, ranging from
family settings to property crime and organized crime, and
firearms as a mechanism for the perpetration of homicide
may in principle be more or less closely associated with a
given situational context. For example, whether a homi-
cide is a crime happening in a domestic setting, as opposed
to a homicide occurring in the context of organized crime,
will probably impact on the likelihood of a firearm having
been used (or not).
FIG. 4
Estimated breakdown of intentional
homicide worldwide, by mechanism
of perpetration, 2017
Source: UNODC Homicide Statistics 2019 (CTS and other official
sources).
FIG. 5
Firearm seizures per firearm homicide, in
comparison with homicide rates, 2016
* Seizures for Australia include cases of an administrative nature.
Note: Countries with known incomplete coverage of seizure data are excluded.
Sources: UNODC Illicit Arms Flows Database (IAFQ and other official sources);
UNODC Homicide Statistics 2019 (CTS and other official sources).
Firearms,
54%
Sharp
object,
22%
Other
mechanism,
24%
0.04
0.2
1
5
25
125
0
0
1
10
100
1,000
Australia*
Slovenia
Spain
Netherlands
Canada
Chile
Serbia
Montenegro
Uruguay
Ecuador
Grenada
Panama
Colombia
Jamaica
Central African Republic
El Salvador
Mexico
Homicide rates per 100,000 population (log scale)
Firearms seized per firearm homicide
(ratio, log scale)
Ratio of firearms seized per firearm homicide (left axis)
Firearms homicide rate per 100,000 (right axis)
Overall homicide rate per 100,000 (right axis)
82
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
situational context. Two such countries were the Nether-
lands and Trinidad and Tobago. Once more, in both cases
the role of firearms appears to be least pronounced in con-
nection with homicides perpetrated by intimate partners
or family members, and much more prominent in con-
nection with organized crime (in the case of the Nether-
lands) and gangs (in the case of Trinidad and Tobago).
Taking into account all four countries, the role of firearms
in robbery-related homicide varied significantly but, by
and large, stayed within the range defined by IPFM homi-
cides on one hand and organized crime or gang-related
homicides on the other.
FIG. 6
Proportion of homicides involving firearms in Canada and El Salvador, by situational context
* Intimate partner or family member.
Source: UNODC homicide statistics (2019).
Some relationships can also be observed between the
mechanism used in the perpetration of homicide and the
gender of homicide victims, as well as homicide perpetra-
tors. While men are a majority of victims of homicide
overall, there are clear differences in the proportion of men
among homicide victims depending on the type of mecha-
nism, with the strongest representation of men being
found among victims of homicides perpetrated by firearm.
Still, this is in a context where men tend to constitute a
majority of victims across all three types of mechanism.
When focusing only on IPFM homicides, however, the
picture is very different. It should be noted that women
0%
20%
40%
60%
80%
100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2008
2009
2010
2011
2012
2008
2009
2010
2011
2012
2008
2009
2010
2011
2012
Gangs Robbery IPFM*
Percentage of homicides involving firearms
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Gangs Organized
crime
Robbery IPFM*
Percentage of homicides involving firearms
Uncertainty range Best estimate
Canada, 2008-2012 El Salvador, 2013
FIG. 7
Estimated distributions* of homicides in the Netherlands by mechanism, according to
situational context, 2007-2015
* UNODC estimates based on independent disaggregations of homicides by mechanism and by situational context and derived using a model which
assumes that the distributions remain stable over time. Available data from the Netherlands did not include a cross-disaggregation of homicides by
mechanism and situational context.
Source: UNODC, elaboration of homicide statistics (2019).
5%
Intimate partner/family member
26%
Robbery
Firearms Sharp object Other mechanism
90%
Organized Crime
33%
Other context
83
Forms of crime linked to firearms 4
FIG. 8
Estimated distributions* of homicides in Trinidad and Tobago by mechanism, according to
situational context, 2006-2015
*UNODC estimates based on independent disaggregations of homicides by mechanism and by situational context and derived using a model which
assumes that the distributions remain stable over time. Available data from Trinidad and Tobago did not include a cross-disaggregation of homicides by
mechanism and situational context.
Source: UNODC, elaboration of homicide statistics (2019).
constitute a majority of victims of IPFM homicides, largely
due to the subset of intimate partner (IP) homicides.
Although data enabling a cross-disaggregation of IPFM
homicides by mechanism and sex are very limited, coun-
try-specific data suggest that, in a context where women
make up the great majority of victims of IPFM homicide,
the proportion of women tends to be consistently higher
among victims of firearm-related IPFM homicides, in
comparison with those IPFM homicides perpetrated by
means of sharp objects.
Indeed, the pattern that emerges from the limited data
suggests that an IPFM homicide is more likely to be per-
petrated using a firearm, and less likely to be perpetrated
by means of a sharp object, when the victim is female.
Considering that the majority of IPFM homicide is IP
homicide, and that the victim and perpetrator are not of
the same sex in the vast majority of IP homicides, this
suggests that men are more likely than women to use a
firearm when killing their female partners, while women
are more likely to resort to a sharp object; this is despite
the “equalizing” effect of a firearm, which renders physical
strength largely irrelevant. The reasons for this pattern are
unclear; one possible explanation could be that men are
more likely to kill with premeditation, while an alternative
explanation could be that women may tend to have less
access to and less familiarity with firearms than men.
4
Civilian holdings
Although firearms account for a significant component of
homicides in most countries worldwide, the level of fire-
arms-related homicide, as well as the relative share (pro-
portion) of firearm-related homicides among homicides
4 See UNODC, Global Study on Homicide, 2019. Booklet 3, Figure 62.
10%
Intimate partner/family member
13%
Robbery
90%
Gang-related
93%
Unknown context
Firearms Sharp object Other mechanism
FIG. 9
Proportion of men among homicide
victims, by type of mechanism,
common range among countries
with available data,* 2015-2016
* 45 countries.
** The common range (interquartile range) represents the “middle half”
of the datapoints; in other words, it excludes data from the lowermost
and topmost quartiles.
Source: UNODC Homicide Statistics (2019).
Average
Median
Common**
(interquartile) range
Firearms
Sharp objects
Other mechanism
84
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
generally, both vary greatly from country to country. The
factors which explain the variation in homicide rates across
countries are multiple and complex, and include socio-
economic characteristics such as inequality as well as the
age composition of a country’s population.
5
Some of these factors can be expected to impact homicide
generally, independently of the mechanism of perpetra-
tion; in contrast, one aspect which may impact firearms-
related homicide specifically is the availability of firearms
in the licit market and their ready accessibility to the gen-
eral population. This aspect is itself complex to measure,
but one metric which may be used is the rate of holdings
of firearms among the civilian population.
6
5 See UNODC, Global Study on Homicide, 2019. Booklet 4.
6 Alternative measures would be, for example, the proportion of indi-
viduals among the population who hold a firearm, or the proportion
of individuals who have access to a firearm within their household.
FIG. 10
Firearms homicide in relation to
civilian holdings of firearms
Sources: UNODC Homicide database (2019); Small Arms Survey.
When this metric is used, alongside other indicators, to
model homicide rates at a global level, its role in explain-
ing homicide-related firearms, although statistically sig-
nificant, appears to be secondary to other factors of a
socioeconomic character, factors which may themselves
be linked, for example, to the underlying level of devel-
opment.
7
However, when considering countries in more
homogenous groups, in particular by distinguishing
between developed and developing countries,
8
the impact
of civilian-held firearms can be observed more clearly.
9
Indeed, considering developed and developing countries
separately, and using estimates of civilian holdings of fire
-
arms published by Small Arms Survey,
10
statistical models
suggest that a 1 per cent increase in the rate of possession
of firearms (number of civilian-held firearms per person)
can be expected to correspond, all other relevant factors
being equal, to an increase of 1.13 per cent
11
in the homi-
cide rate in the case of developing countries (based on 39
countries determined on the basis of data availability) and
0.74 per cent
12
in the case of developed countries (based
on 39 other countries)
13
(see Figure 10).
7 See UNODC, Global Study on Homicide, 2019. Booklet 3, Figure 58.
8 Based on the designation of "developed" and "developing" countries
in the standard country or area codes for statistical use maintained by
the United Nations Statistical Division.
9 See UNODC, Global Study on Homicide, 2019. Booklet 3, page 84.
10
Small Arms Survey, Estimating Global Civilian-held Firearms Numbers.
Briefing Paper, June 2018 (Annex).
11
95 per cent confidence interval: 0.58-1.68 per cent.
12 95 per cent confidence interval: 0.47-1.00 per cent.
13 See UNODC, Global Study on Homicide, 2019. Booklet 3, page 84.
y = 65.4x1.13
R² = 0.318
0.01
0.1
1
10
100
0.001 0.01 0.1 1
Firearm-perpetrated homicide rate,
2013-16 averages (log scale)
Developing countries
0.001
0.01
0.1
1
10
0.001 0.01 0.1 1 10
Firearm-perpetrated homicide rate,
2013-16 averages (log scale)
Civilian-held firearms per person
(log scale)
Developed countries
y = 65.4x
1.13
R² = 0.318
y = 0.97x
0.74
R² = 0.457
85
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
The information and data presented in the Global Study
on Firearms Trafficking provide an insight into the mag-
nitude and extent of firearms trafficking and its links to
other forms of crime. The Study constitutes the result of
the first global data collection exercise via the Illicit Arms
Flows Questionnaire (IAFQ), which builds upon an earlier
effort to shed light on firearms trafficking undertaken by
UNODC in 2015. The following is a discussion of the
main conclusions and policy implications of its findings.
Member States and the international community
have increased their efforts to collect evidence
based information on firearms trafficking, but
more needs to be done to provide the full picture
and stop illicit trafficking flows.
Between 2015 and now, the attention as well as efforts of
Member States and of the international community to
develop, analyse and base policies and strategies for effec-
tive results against firearms trafficking on reliable datasets
on seized and trafficked firearms increased significantly.
This important development, which is also supported by
the commitment of Member States to significantly reduce
illicit arms flows inscribed in Target 16.4 of the Sustain-
able Development Goals, is reflected in various ways.
There is a notable increase of analytical and research prod-
ucts on firearms trafficking, several initiatives by Member
States to enhance their national firearms data collection
capacities, tools and mechanisms, and a considerable
increase in the response rate to the questionnaire by almost
40 per cent compared to the 2015 exercise conducted by
UNODC.
In 80 countries, an amount of over 500,000 firearms have
been seized in each of 2016 and 2017. However, existing
gaps in data coverage in several of these countries and
reported difficulties in detecting illicit trafficking flows
lead to conclude that these amounts are likely to represent
just the tip of the iceberg, as higher amounts of illicitly
circulating firearms remain undiscovered and
unreported.
More efforts are needed to enhance the intelligence picture
on firearms trafficking by looking closer at the criminal
context of each seizure and the whereabouts of the fire-
arms, in order to better understand the nature and extent
of the illicit trafficking flows and devise concrete preven-
tive and control measures to stop the flows and reduce
criminals’ access to these arms.
Seizure data represent one of the most relevant
proxies to disclose and monitor illicit trafficking
flows, but need to be enriched with complemen-
tary analysis on the criminal context and on the
illicit origin of the seized items.
CONCLUSIONS
AND POLICY
IMPLICATIONS
86
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
The study has confirmed the relevance of using seizure
results as an entry point to explore and further analyse the
possible extent and nature of firearms trafficking patterns.
While it is clear that not all seized firearms are necessarily
linked to illicit traffic, seizure information, when usefully
combined with complementary information on the seized
items, the criminal context in which the seizure occurs,
and the whereabout of these items, can help disclose illicit
trafficking routes and patterns.
Firearms trafficking remains a largely
invisible phenomenon
The vast majority of firearms are seized within national
territories, in contexts other than illicit trafficking, either
in the context of another crime, or when circulating within
the national territory. Additional circumstantial informa-
tion, such as the criminal context of the seizure and the
tracing outcome of the items, must be analysed systemati-
cally in order to determine whether these arms were traf-
ficked into the country prior to their seizure.
As a result, illicit firearms trafficking remains most often
invisible and undisclosed, as illicit firearms are hardly
intercepted at their point of diversion, but only when they
re-emerge at the surface in connection with other criminal
activities.
Countries must double their efforts to prevent, detect and
intercept illicit firearms trafficking flows, especially at bor-
ders and during transfers.
Few arms are seized at borders. In order to invert this
trend, countries must intensify their efforts to detect and
combat the illicit trafficking activity itself and prevent that
these firearms make their way in the hands of criminals
and appear again in the context of other crimes.
Enhanced border control and risk profiling capacities are
required to prevent and intercept illicit movement of fire-
arms, including those arriving through less conventional
means such as parcel and postal services deliveries. Detect-
ing and investigating illicit trafficking requires additional
efforts, time and resources, that are often not available to
criminal justice practitioners.
Criminal justice responses tend to underplay the
significance of firearms trafficking by focusing pre-
dominantly on illegal ownership rather than illicit
origin and the criminals involved in trafficking.
The study confirms that, on average, around two thirds
of firearms seizures were based on the legal ground of illicit
possession, while complementary contextual information
and tracing results clearly point to the fact that a consid-
erable portion of these firearms may have been illicitly
trafficked into the country prior to their seizure. The cur-
rent practice of limiting seizure efforts to illicit possession
charges reflect the relative short-sightedness of many crimi-
nal justice systems, that focus more on the firearm as a
tool for crime or as evidence, and do not see the value of
addressing its illicit origin and trafficking. By doing so,
they meet the immediate objective of taking illicit firearms
out of circulation, but the overarching goal of countering
illicit trafficking remains largely undisclosed and
unattended.
Achieving the SDG Target 16.4 requires a fundamental
paradigm shift, and a more strategic vision and proactive
responsiveness from criminal justice practitioners to look
beyond the single firearm and give priority to the objec-
tive of investigating and prosecuting the illicit trafficking
activity, and thus giving visibility and adequate responses
to this underlying crime that is predicate to and fuels all
the subsequent ones.
There is a close link between the domestic and
transnational characteristics of firearms trafficking,
which are not always clearly distinguished by
national authorities. To understand the source of
the issue and conceive appropriate responses,
Member States need to set up systems that enable
them to understand, analyse and react to the
domestic and transnational extent of the issue.
The information and data presented in the study suggest
that, on average, the diversion rate of firearms at domestic
level is high. It is therefore plausible that illicit firearms
circulate within national borders before they are eventually
trafficked abroad. Looking at trafficking that crosses bor-
ders in isolation from domestic trafficking would be mis-
leading because domestic trafficking is often the beginning
of the illicit transnational supply chain of firearms. While
these two phenomena require a joint analysis, distinct
knowledge about the domestic and transnational extent
and characteristics of firearms trafficking is essential for
national authorities and policy makers to understand the
source of the problem and conceive appropriate responses.
However, national legislation is not conducive to grasping
the domestic or transnational nature of the phenomenon.
Only 6 of 53 countries that provided the relevant data
adopted a specific legal concept for transnational firearms
trafficking, rendering the recording of disaggregated data
on firearms seizures difficult.
In order to facilitate distinct data collection efforts on illicit
transnational and domestic arms flows, Member States
should consider developing two distinct legal concepts for
the two phenomena. Keeping close track of seizures by
customs authorities at land border crossings, airports and
harbours on the one hand, and systematically recording,
aggregating and analysing tracing results for seized firearms
on the other, constitute further important measure to help
increasing the understanding of the domestic and trans-
national extent and characteristics of the issue.
87
Forms of crime linked to firearms 4
To prevent and combat firearms trafficking, it is
necessary to scale up capacities, procedures and
tools to identify the illicit origin of seized, found
and surrendered firearms and to record the results
in an accessible manner. Particular emphasis should
be placed on:
- Recording information on firearms and their
criminal context during the different steps of an
investigations in an easily aggregable manner;
- Improving mechanisms and enhancing capacities
for domestic and international tracing of seized
firearms;
- Centralising relevant data in national databases
that can help conduct tracing and support investi-
gation of their possible illicit origin.
The relatively low response rate to the different questions
of the IAFQ and the varying geographical and institutional
coverage of the data in individual countries reflect the lack
of systematic and comprehensive data collection proce-
dures in Member States.
In order to develop a comprehensive insight into firearms
trafficking, Member States should harmonise their record-
ing requirements in the different steps of a firearms-rele-
vant investigation, including the seizure, the in-depth
analysis of the weapon, as well as its tracing outcome. The
particularly low response rate to questions related to trac-
ing results reveal the pressing need for Member States to
enhance mechanisms and capacities for effective domestic
as well as international tracing. Lastly, Member States
should set up centralised databases that can help conduct
tracing and support investigation of their possible illicit
origin. This constitutes a crucial element of any effective
firearms control regime.
Enhancing efforts to detecting and seize parts and
components of firearms, including when shipped
via fast parcel deliveries, may support the fight
against various types of illicit manufacture of
firearms, including conversion, assembly and craft
production.
While the reported overall level of seizures of parts and
components is low in comparison with firearms, the
number of seizures is higher in countries where firearms
in non-factory condition were seized, including artisanally
manufactured, assembled and converted weapons. The
dark number of illicitly circulating parts and components
may likely be higher than the reported numbers, due to
the difficulty in detecting and identifying them as items
belonging to a firearm. Increased attention and capacities
to take illicit parts and components out of circulation,
including when shipped via fast parcel delivery services,
may prevent various types of illicit firearms manufacture,
such as conversion, assembly and craft production.
In order to address transnational firearms traffick-
ing, law enforcement efforts should continue to
target different types of transportation,
in particular vehicles and vessels.
The information and data presented in the study reveal
that vehicles and vessels constitute the primary types of
transportation to traffic firearms across borders. Conse-
quently, continued efforts to detect firearms in vehicles at
land border crossings as well as in vessels at harbours are
required.
Collecting and analysing data on lost and stolen
firearms can further help to develop understand-
ing about firearms trafficking and the point of
diversion of trafficked firearms.
Supplementing the globally adopted indicator to measure
illicit arms flows as one of the Sustainable Development
Goals, Member States and the international community
should further look into systematically collecting and ana-
lysing data on lost and stolen firearms. Such information
can enrich the data on firearms seizures and increase
understanding of illicit arms flows. Data recorded in the
Interpol-administered iARMS database can constitute a
useful starting point in this regard.
While the country of manufacture of seized fire-
arms and the country of diversion often do not
overlap, countries seizing illicitly trafficked fire-
arms should inform the country of manufacture
about the seizure. Such an information exchange is
an important step to improve or put in place pre-
ventive measures at the early stages of the supply
chain and may provide essential information for
the export risk assessment to be undertaken by the
exporting country.
An important share of trafficked firearms originate as
legally manufactured firearms that are diverted to the black
market only after their legal export. However, under inter-
national and regional instruments, such as the Arms Trade
Treaty or the Central African Convention against the Fire-
arms, their Parts and Components and Ammunition,
countries are obliged to assess the risk of future diversion
during their export risk assessment. While post-delivery
verifications may constitute an effective measure to sup-
port the export risk assessment, information on trafficked
firearms from the seizing country may constitute another
important source of information for the manufacturing
country. Member States should therefore maintain open
communication channels and information exchange on
trafficked and diverted firearms.
Integrated and comprehensive data collection
mechanisms and use of standardized templates
and tools can help improve the situation. Without
this, Member States risk to miss attainment and
monitoring of SDG Target 16.4. Increased participa-
tion and support for global data collection efforts
is also required.
88
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
The low priority given by criminal justice systems to the
issue of firearms trafficking is often also reflected in the
rudimentary and isolated way in which data on firearms
are sometimes collected and recorded in some countries.
Consequently, data collection mechanisms and tools of
many countries are also often not sufficiently equipped to
fully capture the phenomenon of illicit trafficking and
related offences, including new or emerging threats like
trafficking in parts and components, parcel service and
internet-enabled trafficking; conversion and modification
of arms into live firing weapons, etc.
With increased visibility and awareness on the importance
of addressing firearms trafficking, more priority should
also be given to ensuring integrated and comprehensive
methods of gathering, analysing and using data on fire-
arms. Countries need to recognize the relevance of look-
ing beyond firearms as an instrument or evidence, into
their origin and possible trafficking, and to adapt their
data collection and analysis mechanism accordingly in
order to integrate broader crime data into the overall intel-
ligence picture surrounding illicit firearms.
There is a need to enhance the intelligence picture on
firearms trafficking by enhancing their counting mecha-
nism from the bottom up, in order to facilitate the gath-
ering and analysis of circumstantial information such as
the criminal context, links to specific security threats like
terrorism or organized crime, and to support evidence-
based intelligence analysis and strategic decision-making.
Development and use of standardized tools, such as tem-
plates and automated applications that can help disaggre-
gate the information from the moment it is produced /
generated, and reduce risks of errors, can greatly contribute
to enhance the picture and the analysis
Poor tracing practice and insufficient efforts to
report on them reduce the likelihood to disclose
and counter illicit arms flows, and put Member
States at risk of not attaining and monitoring the
SDG Target 16.4. Increased participation and sup-
port for global data collection efforts is required.
While seizure data provide the closest proxy to monitor
firearms trafficking, only tracing the illicit origin can help
establish the point of diversion of the seized firearms, and
identify trafficking routes and patterns.
Countries must systematically trace and report on their
efforts to establish the illicit origin of firearms seized in
crime, with a view to investigate and prosecute the illicit
trafficking offence, beyond and regardless of the main
offence that has given rise to the law enforcement action
leading to the seizure of the firearm. Without this, coun-
tries will not be able to detect and disrupt illicit trafficking
networks and reduce the illicit arms flows by 2030.
Yet, looking at the overall number of countries providing
processable data on the SDG-related question (14 coun-
tries) and the number of seized, found and surrendered
arms they were able to trace to their illicit origin (28 per
cent on average), Member States are at risk to fail attain-
ing and monitoring the attainment of SDG Target 16.4.
It is therefore necessary to scale up capacities, procedures
and tools to identify the illicit origin of seized, found and
surrendered firearms and to record the results in an acces-
sible manner.
Transnational firearms trafficking constitutes a
complex crime, with often fatal consequences.
To counter the phenomenon effectively, Member
States and the international community should
invest in further research into existing links to
other security threats and develop appropriate,
multi-sectorial approaches.
The study confirms the interconnectedness of seized fire-
arms on the one hand and drug trafficking as well as armed
violence, in particular homicide, on the other. Further
analysis on these links as well as on the interconnectedness
between illicit firearms and other substantive topics, such
as trafficking in other commodities and terrorism should
be conducted. Insights into the interplay between illicit
firearms on the one hand and corruption and money laun-
dering on the other may further help in enhancing inves-
tigative approaches against these crimes.
While the study confirms that firearms used in homicide
have a particular gender dimension, the impact of illicit
firearms on men and women, boys and girls should be
further researched and addressed.
89
GLOBAL STUDY ON FIREARMS TRAFFICKING
2020
ANNEX
REGIONAL SUMMARY
AND TRENDS
Africa
In Africa, the largest quantities of seized weapons seized
were registered in Angola and Kenya. Aside from the prev-
alence of shotguns generally, notable proportions of
machine guns were seized in Tunisia and of submachine
guns in Burundi. In Central African Republic, aside from
miscellaneous weapons such as grenades, artisanal weapons
and cannons, the remaining seized weapons were predomi-
nantly rifles and submachine guns.
However, many countries in Africa and Asia appear to
have a lower capacity to intercept and report trafficked
firearms, which may lead to underreporting of some types
of firearms. Moreover, the total figures reported by coun-
tries include seizures which are not directly connected to
trafficking. Based on customs seizures at borders, rifles
emerge at par with pistols. This suggests that firearms such
as rifles may play a bigger role in global trafficking pat-
terns than what is reflected in the currently available data.
Looking more closely, links emerge between trafficking
patterns and broader regional contexts. For example, coun-
tries with higher levels of violent deaths and homicide,
particularly in Africa and Latin America and the Carib-
bean, tend to seize a higher percentage of firearms con-
nected to violent crime. Similarly, in countries with higher
levels of drug trafficking, more arms are seized linked to
that activity.
FIG. 1
Average distribution* of reported seized arms in Africa, by type, 2016-17
* Simple average based on data for 18 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms or
SALWs.
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Sources: UNODC IAFQ and other official sources.
Other undistinguished weapons
(unspecified/aggregate)**
Pneumatic, blank-firing and gas
weapons***
Weapons or related items
other than firearms/SALWs,
including:***
Unclassified/
unknown
Pistols,
28%
Rifles,
22%
Revolvers,
2%
Submachine guns, 8%
Machine guns, 2%
Other SALWs
Shotguns,
38%
Firearms/
SALWs,
68%
90
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 2
Distribution of seized arms by type, countries in Africa, 2016-17
* Includes other firearms, other SALWs and other weapons. Also includes weapons reported under “Other” without sufficient information to allow
further classification or disaggregation; some of these weapons may be firearms of the six foregoing standard types.
** Pneumatic, blank-firing and gas weapons fall under the preceding designation “Weapons other than firearms/SALWs”, but are distinguished
whenever the available data allows.
Sources: UNODC IAFQ and other official sources.
0%
Pistols
100% 0%
Revol-
vers
60% 0%
Rifles
70% 0%
Shotguns
100% 0%
Machine
guns
60% 0%
Submachine
guns
100% 0%
Other*
90% 0%
Unknown/
unclassified
100%
Algeria
Angola
Botswana
Burkina Faso
Burundi
Cabo Verde
Cameroon
Central African Republic
Cote d'Ivoire
Ghana
Guinea
Kenya
Libya
Mauritius
Morocco
Sudan
Togo
Tunisia
Percentage
Pistols Revolvers Rifles Shotguns
Machine guns Submachine guns Other firearms Other SALWs
Other arms/weapons
(unspecified/aggregate)
Weapons other
than firearms/SALWs
Pneumatic, blank-firing
and gas weapons**
Unknown/
unclassified
91
Annex
FIG. 3
Arms seized by countries in Africa, by type, 2016-17 (9 countries with largest quantities seized)
Sources: UNODC IAFQ and other official sources.
FIG. 4
Arms seized by countries in Africa, by type, 2016-17 (9 countries with lowest quantities seized)
* Data for Ghana were available for 2016 only; data for Cameroon and Guinea were available for 2017 only.
Sources: UNODC IAFQ and other official sources.
0
400
800
1,200
1,600
2,000
2,400
0
2,000
4,000
6,000
8,000
10,000
12,000
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Angola
(left axis)
Kenya
(left axis)
Sudan
(right axis)
Tunisia
(right axis)
Algeria
(right axis)
Burkina
Faso
(right axis)
Central
African
Republic
(right axis)
Libya
(right axis)
Cote
d'Ivoire
(right axis)
Number of arms (all countries except Angola and Kenya)
N
um
b
er o
f
arms
(A
ngo
l
a an
d
K
enya
)
0
50
100
150
200
250
300
2016
2017
2016
2017
2016
2017
2016
2017
2017
2016
2017
2017
2016
2017
2016
Cabo Verde Morocco Burundi Togo Cameroon*Botswana Guinea* Mauritius Ghana*
Number of arms
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
92
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 5
Average distribution* of reported seized arms in the Americas, 2016-17
* Simple average based on data for 26 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms or
SALWs.
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Sources: UNODC IAFQ and other official sources.
Americas
In Latin America and the Caribbean, the largest quantities
of seized weapons were reported by Colombia and Argen-
tina. However, comprehensive data for Brazil were not
available. Excluding weapons other than firearms/SALWs,
as well as any firearms/SALWs which could not be classi-
fied and quantified into the respective category, the Ameri-
cas were the region with the highest proportions of pistols
(on average, 52 per cent) and revolvers (24 per cent). The
proportion of rifles seized in Mexico was significantly
higher than in other countries in Latin America and the
Caribbean.
The data collection exercise carried out for the present
study, which was focused on data for the years 2016 and
2017, built on an earlier exercise carried out by UNODC.
1
Hence, for some countries, seizure data were also available
for earlier years. Despite a gap in the reporting periods
and the potential for issues of comparability, in some cases
it was also possible to derive meaningful longer-term
increasing or decreasing trends. In Latin America, this was
possible for three countries, all of which exhibited clear
decreasing trends.
1 UNODC Study on Firearms, 2015.
Other undistinguished weapons
(unspecified/aggregate)**
Pneumatic, blank-firing
and gas weapons***
Unclassified/
unknown
Pistols,
52%
Rifles, 8%
Revolvers,
24%
Submachine guns, 1%
Machine guns
Other firearms, 1%
Other SALWs
Shotguns,
14%
Firearms/
SALWs,
79%
93
Annex
FIG. 6
Distribution of seized arms by type, countries in the Americas, 2016-17
* Includes other firearms, other SALWs and other weapons. Also includes weapons reported under “Other” without sufficient information to allow
further classification or disaggregation; some of these weapons may be firearms of the six foregoing standard types.
** Pneumatic, blank-firing and gas weapons fall under the preceding designation “Weapons other than firearms/SALWs”, but are distinguished
whenever the available data allows.
Sources: UNODC IAFQ and other official sources.
0%
Pistols
100% 0%
Revol-
vers
70% 0%
Rifles
60% 0%
Shotguns
60% 0%
Machine
guns
60% 0%
Sub-
machine
guns
60% 0%
Other*
60% 0%
Unknown/
unclassified
100%
Antigua and Barbuda
Argentina
Bahamas
Bolivia (Plurinational State of)
Brazil
Canada
Chile
Colombia
Costa Rica
Cuba
Dominican Republic
Ecuador
El Salvador
Grenada
Guatemala
Guyana
Honduras
Jamaica
Mexico
Panama
Paraguay
Peru
Saint Lucia
Suriname
United States of America
Uruguay
Percentage
Pistols Revolvers Rifles Shotguns
Machine guns Submachine guns Other firearms Other SALWs
Other arms/weapons
(unspecified/aggregate)
Weapons other
than firearms/SALWs
Pneumatic, blank-firing
and gas weapons**
Unknown/
unclassified
94
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 8
Arms seized by countries in the Americas, by type, 2016-17 (9 countries with intermediate
quantities seized)
* Includes only seizures in the records of the Federal Police Tracing Centre of Brazil.
Sources: UNODC IAFQ and other official sources.
0
1,000
2,000
3,000
4,000
5,000
6,000
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Uruguay Brazil* Peru El
Salvador
Ecuador Panama Dominican
Republic
Jamaica Bahamas
Number of arms
FIG. 7
Arms seized by countries in the Americas, by type, 2016-17 (9 countries with largest
quantities seized)
* Includes firearms submitted for tracing to the Bureau of Alcohol, Tobacco, Firearms and Explosives by a law enforcement agency, resulting from seizure
as well as abandonment, buy-back program, or other recovery method. Moreover, only seized firearms submitted for tracing are included. Firearms
submitted for tracing after recovery do not represent the entire set of all seized firearms.
¹ Data for Canada were available for 2016 only.
Sources: UNODC IAFQ and other official sources.
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
2016
2017
2016
2017
2016
2017
2016
2017
2016
2016
2017
2016
2017
2016
2017
2016
2017
United States
of America*
(left axis)
Colombia
(right axis)
Argentina
(right axis)
Costa
Rica
(right axis)
Canada¹
(right
axis)
Mexico
(right axis)
Chile
(right axis)
Guatemala
(right axis)
Honduras
(right axis)
Number of arms
(United States of America only)
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
Number of arms
(all countries except United States of America)
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
95
Annex
FIG. 9
Arms seized by countries in the Americas, by type, 2016-17 (9 countries with lowest quantities
seized)
*Data for Paraguay were available for 2016 only.
Sources: UNODC IAFQ and other official sources.
0
20
40
60
80
100
120
140
160
180
200
2016
2017
2016
2017
2016
2017
2016
2017
2016
2016
2017
2016
2017
2016
2017
Suriname Bolivia
(Plurinational
State of)
Guyana Saint
Lucia
Paraguay*Antigua
and
Barbuda
Grenada Cuba
Number of arms
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
FIG. 10
Significant* longer-term trends in the number of arms seized by countries in
Latin America, 2010-2017
* Due to a revised data collection instrument, data prior to 2016 are not guaranteed to be comparable to data for 2016-17, with the exception of
updated historical data. This graph shows only countries for which at least 5 datapoints were available over the period 2011-2017 and the correspond-
ing simple linear regression model yielded p-values that were significant at the 90% level.
** For 2014 and 2015, no data were available for Mexico and for Panama.
Sources: UNODC IAFQ and other official sources.
-
1,000
2,000
3,000
4,000
5,000
-
10,000
20,000
30,000
40,000
50,000
2010
2011
2012
2013
2014**
2015**
2016
2017
Number of arms seized (Panama)
Number of arms seized (Colombia and Mexico)
Colombia
Mexico
Panama (right axis)
Colombia, line of best fit
Mexico, line of best fit
Panama, line of best fit
Missing data
96
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Europe
Among all regions, the types of weapons seized in Europe
were the most diversified. Among weapons which could
be clearly classified as firearms and SALWs, pistols
accounted on average for 35 per cent of seizures, followed
by rifles (27 per cent), shotguns (22 per cent) and revolv-
ers (11 per cent), with submachine guns and machine guns
accounting for most of the remainder (4.5 per cent).
2
Moreover, on average Europe registered the highest pro-
portions of miscellaneous types of weapons – this may be
related to the prominence of conversion and other forms
of improvisation as ways to obtain illicit firearms.
As for the Americas, in the case of Europe it was also pos-
sible to identify some statistically meaningful longer-term
decreasing or increasing trends, namely in the cases of
Spain, Lithuania and Romania.
Historical data were also available for countries in the
Western Balkans; however, in this case, no statistically sig-
nificant increasing or decreasing trends were identified.
2 This breakdown excludes weapons other than firearms/SALWs, as well
as any firearms/SALWs which could not be quantified and classified
into the respective category.
FIG. 11
Average distribution* of reported seized arms in Europe, 2016-17
* Simple average based on data for 26 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms
or SALWs.
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Sources: UNODC IAFQ and other official sources.
Other undistinguished weapons
(unspecified/aggregate)**
Pneumatic, blank-firing
and gas weapons***
Weapons or related items
other than firearms/SALWs,
including:***
Unclassified/
unknown
Pistols,
35%
Rifles,
27%
Revolvers,
11%
Submachine
guns, 3%
Machine guns, 2%
Other SALWs
Shotguns,
22%
Firearms/
SALWs,
62%
97
Annex
FIG. 12
Distribution of seized arms by type, countries in Europe, 2016-17
* Includes other firearms, other SALWs and other weapons. Also includes weapons reported under “Other” without sufficient information to allow
further classification or disaggregation; some of these weapons may be firearms of the six foregoing standard types.
** Pneumatic, blank-firing and gas weapons fall under the preceding designation “Weapons other than firearms/SALWs”, but are distinguished
whenever the available data allows.
Sources: UNODC IAFQ and other official sources.
0%
Pistols
60% 0%
Revol-
vers
60% 0%
Rifles
60% 0%
Shotguns
90% 0%
Machine
guns
60% 0%
Sub-
machine
guns
60% 0%
Other*
90% 0%
Unknown/
unclassified
100%
Albania
Belarus
Belgium
Croatia
Denmark
Greece
Hungary
Italy
Liechtenstein
Lithuania
Luxembourg
Montenegro
Netherlands
North Macedonia
Norway
Portugal
Republic of Moldova
Romania
Russian Federation
Serbia
Slovakia
Slovenia
Spain
Sweden
Ukraine
United Kingdom
Percentage
Pistols Revolvers Rifles Shotguns
Machine guns Submachine guns Other firearms Other SALWs
Other arms/weapons
(unspecified/aggregate)
Weapons other
than firearms/SALWs
Pneumatic, blank-firing
and gas weapons**
Unknown/
unclassified
98
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 13
Arms seized by countries in Europe, by type, 2016-17 (9 countries with largest quantities seized)
Sources: UNODC IAFQ and other official sources.
FIG. 14
Arms seized by countries in Europe, by type, 2016-17 (9 countries with intermediate quantities
seized)
Sources: UNODC IAFQ and other official sources.
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Russian
Federation
Spain Belarus Portugal Greece Italy Netherlands Belgium Ukraine
Number of arms
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Croatia Serbia Slovakia United
Kingdom
Sweden Denmark Luxembourg Albania Norway
Number of arms
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
99
Annex
FIG. 15
Arms seized by countries in Europe, by type, 2016-17 (8 countries with lowest quantities seized)
Sources: UNODC IAFQ and other official sources.
0
50
100
150
200
250
300
350
400
450
500
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
2016
2017
Montenegro Slovenia Lithuania Hungary North
Macedonia
Romania Republic
of Moldova
Liechtenstein
Number of arms
FIG. 16
Significant* longer-term trends in the
number of arms seized by countries
in Europe, 2010-2017
* Due to a revised data collection instrument, data prior to 2016 are not
guaranteed to be comparable to data for 2016-17, with the exception of
updated historical data. This graph shows only countries for which at least
5 datapoints were available over the period 2011-2017 and the corre-
sponding simple linear regression model yielded p-values which were sig-
nificant at the 90% level. The Russian Federation was also excluded due to
the fact that data prior 2016 covered customs seizures only.
** For 2014 and 2015, no data was available for Lithuania and Romania.
Sources: UNODC IAFQ and other official sources.
FIG. 17
Seizures of arms in the Western Balkans,
2010-2017
Note: Due to a revised data collection instrument, data prior to 2016 are not
guaranteed to be comparable to data for 2016-17. Therefore, the comparison
needs to be made with caution.
Sources: UNODC IAFQ and other official sources.
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
11,000
12,000
0
100
200
300
400
500
600
2008
2009
2010
2011
2012
2013
2014**
2015**
2016
2017
Number of arms seized (Spain)
Number of arms seized (Lithuania and Romania)
Lithuania Romania
Missing data Spain (right axis)
Lithuania, line of best fit Romania, line of best fit
Spain, line of best fit
0
500
1,000
1,500
2,000
2010
2011
2012
2013
2014
2015
2016
2017
Nmber of arms seized
Albania Montenegro
North Macedonia Serbia
Missing data
100
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
FIG. 18
Average distribution* of reported seized arms in Asia, 2016-17
* Simple average based on data for 10 countries.
** Includes weapons reported under “Other” without sufficient information to allow further classification; some of these weapons may be firearms or
SALWs.
*** For some countries, the reported seizure data included weapons other than firearms/SALWs; however, data on such weapons were not explicitly
requested by the questionnaire. Hence the share of such weapons is subject to variations in the reporting practice across countries.
Sources: UNODC Illicit Arms Flows Database (IAFQ and other official sources.)
FIG. 19
Distribution of seized arms by type, countries in the Asia and Oceania, 2016-17
* Includes other firearms, other SALWs and other weapons. Also includes weapons reported under “Other” without sufficient information to allow
further classification or disaggregation; some of these weapons may be firearms of the six foregoing standard types.
** Pneumatic, blank-firing and gas weapons fall under the preceding designation “Weapons other than firearms/SALWs”, but are distinguished
whenever the available data allows.
Sources: UNODC Illicit Arms Flows Database (IAFQ and other official sources.)
Other undistinguished
weapons
(unspecified/aggregate)**
Pneumatic, blank-firing
and gas weapons***
Unclassified/
unknown
Pistols,
34%
Rifles,
16%
Revolvers,
10%
Submachine
guns, 1%
Machine
guns, 1%
Shotguns,
37%
Firearms/
SALWs,
82%
0%
Pistols
60% 0%
Revol-
vers
60% 0%
Rifles
60% 0%
Shotguns
80% 0%
Machine
guns
60% 0%
Sub-
machine
guns
60% 0%
Other*
60% 0%
Unknown/
unclassified
100%
Australia
Azerbaijan
Japan
Kazakhstan
Kyrgyzstan
Lebanon
Myanmar
Nepal
Philippines
Qatar
Tajikistan
Percentage
Asia and Oceania
In Asia and Oceania, coverage was rather limited; hence
representative patterns cannot be reliably elucidated. The
number of seized weapons was very high in Australia, but
this included seizures of an administrative nature. The
relative importance of shotguns was similar to that in
Africa, with relatively high proportions of this type of
weapon being registered in the neighbouring countries of
Kazakhstan, Kyrgyzstan and Tajikistan, as well as
Lebanon.
Pistols Revolvers Rifles Shotguns
Machine guns Submachine guns Other firearms Other SALWs
Other arms/weapons
(unspecified/aggregate)
Weapons other
than firearms/SALWs
Pneumatic, blank-firing
and gas weapons**
Unknown/
unclassified
101
Annex
FIG. 20
Arms seized by countries in Asia and Oceania, by type, 2016-17 (6 countries with largest
quantities seized)
Sources: UNODC Illicit Arms Flows Database (IAFQ and other official sources.)
FIG. 21
Arms seized by countries in Asia, by type, 2016-17 (5 countries with lowest quantities seized)
*Data for Kyrgyzstan were available for 2017 only.
Sources: UNODC Illicit Arms Flows Database (IAFQ and other official sources.)
0
10
20
30
40
50
60
70
80
90
2016 2017 2017 2016 2017 2016 2017 2016 2017
Qatar Kyrgyz-
stan*
Nepal Tajikistan Philippines
Number of arms
0
500
1,000
1,500
2,000
2,500
3,000
0
5,000
10,000
15,000
20,000
25,000
30,000
2016 2017 2016 2017 2016 2017 2016 2017 2016 2017 2016 2017
Australia
(left axis)
Kazakhstan
(left axis)
Lebanon
(right axis)
Japan
(right axis)
Azerbaijan
(right axis)
Myanmar
(right axis)
Number of arms (all countries except Australia and Kazakhstan)
Number of arms (Australia and Kazakhstan only)
Unknown/unclassified
Other weapons/related items
Weapons or related items other than firearms/SALWs (undistinguished)
Pneumatic, blank-firing and gas weapons
Other arms/weapons (unspecified/aggregate)
Other SALWs
Other firearms
Submachine guns
Shotguns
Rifles
Revolvers
Pistols
Machine guns
102
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Types of firearms
Types of Weapons Example
“Machine gun”: Firearm that automatically shoots more than once without manual reloading, by a
single function of the trigger. A machine gun continues to load and fire ammunition until the trigger,
or other activating device, is released, the ammunition is exhausted, or is jammed. Machine guns can
have automatic firing systems, but often can be set to fire in semi-automatic mode.
“Pistol”: Firearm designed for semi-automatic operation. The chamber is part of the barrel. Cartridges
are generally loaded into an ammunition magazine which is inserted into the grip. As long as car-
tridges are present in the ammunition magazine and the firearm is functioning properly, the action of
the pistol is responsible for the feeding and chambering of the cartridge and the extraction and ejec-
tion of the cartridge case once the cartridge has been fired. The firing systems of pistols can be trig-
ger, repetitive, semi-automatic or automatic. Types of pistols are the single shot pistol, repeating pistol,
semi-automatic pistol and automatic pistol.
“Revolver”: A short or hand-held firearm which has a revolving cylinder with a number of chambers.
These chambers are designed to be manually loaded with cartridges of the appropriate calibre; then,
as the cylinder rotates into position under the hammer, the trigger can be pulled, releasing the
hammer which causes the cartridge to be fired. According to the system of operation of the trigger
tail, revolvers can be of double or single action.
“Rifle (including carbine)”: A relatively long-barrelled firearm, fired from the shoulder, with a series of
spiral grooves cut inside the barrel (referred to as rifling) imparting spin to the projectile. Some rifles
have a detachable magazine similar to the pistols described above, and others have integral maga-
zines. A carbine resembles a rifle but has a shorter barrel. Rifles or carbines have single-shot, repeat-
ing, semi-automatic or fully automatic systems.
“Short shotgun (pistolon)”: A handgun with one or more smoothbore barrels. Short shotguns have
single-shot systems.
“Shotgun”: A shoulder-fired long gun with one or two unrifled barrels, usually designed to shoot a
large number of small projectiles (referred to as shot) rather than a single large projectile (referred to
as bullet). The firing systems of shotguns can be single-shot, repeating or semi-automatic.
Sub-machine gun”: A hand-held, lightweight, short-barrelled machine gun consisting of relatively
low-energy handgun-type cartridges and fired from the hand, hip or shoulder. Submachine guns have
semi-automatic firing systems. (Where the firing system is automatic, firearms are classified as auto-
matic pistols or automatic machine guns.)
103
Annex
Types of Parts and Components Example
“Barrel”: A metal tube, through which a projectile (or shot charge) travels under the force of
the charge out of its front end. The barrel is fixed to the receiver/frame.
“Bolt”: A mechanical part of a firearm (mostly semi-automatic pistols) that blocks the rear
chamber while firing, but moves aside to allow another cartridge to be inserted.
“Breech block”: The movable part of the firing system that seals the moment of firing,
preventing gases from escaping. Most modern small firearms use a bolt.
“Cylinder”: A cylindrical, rotating part of a revolver that contains multiple cartridge chambers.
“Frame or receiver”: The main element of a hand gun to which the barrel and the stock are
attached. The receiver holds other parts and components, such as the breech mechanism, trig-
ger and firing mechanism.
“Magazine”: The ammunition storage and feeding device of a firearm within or attached to
a repeating firearm. Magazines can be removable or integral to the firearm.
“Silencer”: The part designed to reduce the sound and the visible muzzle flash generated by
firing, by slowing down the escaping propellant gas or reducing the velocity of the bullet.
“Slide”: The part that generally houses the firing pin and the extractor, and serves as the
bolt for the majority of semi-automatic pistols. It is spring-loaded to chamber a fresh cartridge
provided that the magazine is not empty.
104
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Glossary
Term Definition
Altered markings
Markings on a weapon that were modified, damaged, or removed in order to make the firearm
unidentifiable.
Ammunition
The complete round or its components, including cartridge cases, primers, propellant powder,
bullets or projectiles used in firearms.
Arms
Weapons reported as seized, found or surrendered through the Illicit Arms Flow Questionnaire.
The target universe is that of all categories of firearms and of small arms and light weapons;
however in some cases the reported figures may include other weapons.
Assembled arms
Any arms illegally put together from multiple industrially manufactured parts and components,
without being part of an industrial process.
Land Border
The customs office between two jurisdictions, but also to border crossings or checkpoints when a
border has no customs office.
Case/instance
A single act of seizing arms, their parts and components or ammunition, regardless of how many
items are seized collectively. For example, the seizure of a shipment containing 200 arms and 500
rounds of ammunition being trafficked illicitly via sea one case or instance, despite the fact that 700
items are being apprehended.
Converted arms
Any arms such as blank firing, air-soft, paintball or air cartridge that is adapted or modified to
enable a shot, bullet or other projectile to be discharged that is capable of lethal injury.
Country of manufacture
The country where the item was manufactured. In the case of assembled arms, the country where
the weapon was assembled should be considered.
Departure country
The country from where the traffickers originally departed before entering your national territory, or
from where the arms were shipped, in the case of unaccompanied shipments. If such information is
not available, please consider the country where the item was last legally recorded or, in cases where
such record does not exist, the country of departure established through intelligence.
National/within national territory
Seizures that occur in national territory, but do not take place in airports, seaports or harbours, or
in land borders, as defined above.
Firearm
Any portable barrelled weapon that expels, is designed to expel or may be readily converted to expel
a shot, bullet or projectile by the action of an explosive, excluding antique firearms or their replicas.
Antique firearms and their replicas shall be defined in accordance with domestic law. In no case,
however, shall antique firearms include firearms manufactured after 1899.
Found item
Any weapons, parts and components or ammunition apprehended by authorities not linked to an
intentional or planned investigation or inspection, neither attributable to any apparent holder or
owner, regardless of whether the item was reported lost or stolen.
Illicit manufacture
The manufacturing or assembly of arms, their parts and components, or ammunition:
(a) From parts and components illicitly trafficked;
(b) Without a licence or authorization from a competent authority of the State party where the
manufacture or assembly takes place; or
© Without marking the weapon at the time of manufacture with a unique marking providing the
name of the manufacturer, the country or place of manufacture and the serial number; or maintain-
ing any alternative unique user-friendly marking with simple geometric symbols in combination with
a numeric and/or alphanumeric code, permitting ready identification by all States of the country of
manufacture. The manufacture of parts and components must be licensed and authorized in accord-
ance with national law.
“Illicit manufacturing” also refers to illicitly reactivated arms. For example, reactivating a deactivated
weapon, which is not considered to be a weapon anymore, is considered illicit manufacture, as well
as manufacturing a weapon at home with parts and components trafficked through the web with-
out licence or authorization.
Illicit possession
The unlawful possession of regulated or prohibited firearms, their parts and components or
ammunition. Carrying a handgun without a license is an example of illicit possession.
Illicit trafficking
The import, export, trans-shipment, acquisition, sale, delivery, movement or transfer of arms, their
parts and components, and ammunition from or across the territory of one State to that of another
State if any one of the States concerned does not authorize it in accordance with national law of
one of the countries involved or if the arms are not uniquely marked upon manufacture and marked
upon import. Arms that were not properly deactivated according to the national legislation of the
destination country can also be illicitly trafficked or smuggled.
105
Glossary
Term Definition
Illicit use
This is the unlawful use of regulated or prohibited firearms, their parts and components or
ammunition. The illegal discharge of arms is an example of illicit use.
Individuals convicted
Persons found guilty by any legal body authorized to pronounce a conviction under national
criminal law, whether or not the conviction was later upheld.
Individuals prosecuted
Alleged offenders against whom prosecution commenced in the reporting year(s). Persons may be
prosecuted by the public prosecutor or the law enforcement agency responsible for prosecution.
All persons for which prosecution starts should be counted, irrespective of the case-ending decision
Intended destination
The country for which the arms, parts and components, and ammunition were destined.
Your country may also be considered in this category.
Last legal record
The last officially recorded information available about the item, its status (deactivated, stolen, lost,
seized, found, surrendered, sent for destruction, confiscated, in transit, etc.) and its legal end-user.
The identification of the last legal record may require the initiation of several individual tracing
requests.
Make
The name or brand of the manufacturer of a given weapon. Examples include: Beretta, Colt, Glock,
Smith and Wesson, and Taurus.
Modified arms
Any arms modified to increase its efficiency or damage capacities. For example, increased capacity
of the magazine, cut barrel, modified from semi-automatic to automatic.
Natural resources A naturally occurring source of wealth. For example, plants, mineral deposits, water, etc.
Parts and components
Any element or replacement element specifically designed for a firearm and essential to its opera-
tion, including a barrel, frame or receiver, slide or cylinder, bolt or breech block, and any device
designed or adapted to diminish the sound caused by firing a firearm.
Point of diversion
The point in space and time, and/or circumstances when arms, parts and components, and ammu-
nition left the licit circuit and entered the illicit one. For example, a weapon was manufactured in
country A, then exported to country B, where it was stolen, and later trafficked to country C,
where it was seized. In this scenario, the point of diversion occurred when the weapon was stolen
in country B. Identifying the point of diversion entails tracing the last available legal record of the
weapon in country B, either the last legal owner in the country or the record that indicates it was
stolen.
Reactivated arms
Any weapon previously deactivated by the competent authority with a view to rendering it
permanently inoperable and that has illegally been returned to working condition.
Seized items
Arms, their parts and components and ammunition that have been temporarily physically
apprehended by a competent authority in relation to a suspected criminal offence or administrative
violation related to these items. Seized items can be permanently confiscated at a later stage by the
state, or returned to their legitimate owners.
Seizures related to administrative
offences
Items seized in circumstances involving an action prohibited by a national criminal code or any
offence listed in this questionnaire, including, among others, trafficking and smuggling of arms.
The item can be the main commodity, the instrument, or derived from the offence.
Small Arms and Light Weapons
(SALWs)
Any man-portable lethal weapon that expels or launches, is designed to expel or launch, or may be
readily converted to expel or launch a shot, bullet or projectile by the action of an explosive, exclud-
ing antique small arms and light weapons or their replicas. Antique small arms and light weapons
and their replicas will be defined in accordance with domestic law. In no case will antique small
arms and light weapons include those manufactured after 1899:
(a) “Small arms” are, broadly speaking, weapons designed for individual use. They include, inter
alia, revolvers and self-loading pistols, rifles and carbines, sub-machine guns, assault rifles and
light machine guns;
(b) “Light weapons” are, broadly speaking, weapons designed for use by two or three persons
serving as a crew, although some may be carried and used by a single person. They include,
inter alia, heavy machine guns, hand-held under-barrel and mounted grenade launchers, port-
able anti-aircraft guns, portable anti-tank guns, recoilless rifles, portable launchers of anti-tank
missile and rocket systems, portable launchers of anti-aircraft.
106
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Term Definition
Surrendered item
Any arms, their parts and components, and ammunition willingly handed to authorities, that is not
linked to a planned investigation or inspection. The surrender may occur as a personal initiative of a
citizen, in the context of a voluntary surrender campaign, and disarmament, demobilisation and
reintegration processes, inter alia.
Tracing
The systematic tracking of arms and, where possible, their parts and components, and ammunition,
at a national and/or international level for the purpose of assisting the competent authorities of
States Parties in detecting, investigating and analysing illicit manufacturing and illicit trafficking.
Searching the national record for example is considered as tracing.
Transit country
The country through which the arms, their parts and components, and ammunition transited before
reaching the destination country. There could be more than one transit country between the depar-
ture/source and destination country.
Typical price in the licit market
The typical price paid at a licit firearms specialized business for a specific weapon. This may refer to
the median (or alternatively, the average) price derived from several observations obtained through a
statistical data collection effort, or it may be obtained from knowledge derived from professional
expertise in the field.
Typical price in the street / black
market
The typical price paid for a specific weapon in the street or black market. This may refer to the
median (or alternatively, the average) price derived from several observations obtained through
intelligence operations (e.g., undercover initiatives), or it may be obtained from knowledge derived
from professional expertise in the field.
Uniquely marked/uniquely
identifiable
A uniquely marked item has a unique marking providing the name of the manufacturer, the coun-
try or place of manufacture and the serial number, or maintain any alternative unique user-friendly
marking with simple geometric symbols in combination with a numeric and/or alphanumeric code,
permitting ready identification by all States of the country of manufacture.
Violent crime
A crime in which the perpetrator uses or threatens to use force upon a victim. Examples include
homicide, intended homicide, rape, among others.
107
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Africa
Eastern Africa
Burundi
2016
IAFQ 3 11 0 0 0 179 0 0 193
2017
IAFQ 0 4 0 0 0 38 0 0 42
Kenya
2016
IAFQ 31 769 156 3699 63 94 28
1
424 5264
2017
IAFQ 51 738 840 5720 2136 202 41
1
0 9728
Mauritius
2016
IAFQ 0 0 0 2 1 0 18
2
021
2017
IAFQ 0 3 1 3 0 0 10
3
017
Middle Africa
Angola
2016
IAFQ 11475
2017
IAFQ 10195
Cameroon
2017
IAFQ 0 43 0 37 0 0 30 0 110
Central African Republic
2016 UNPOL 0 6 0 32 0 3 263 0 304
2017 UNPOL 0 5 0 57 0 17 132 0 211
Northern Africa
Algeria
2016
IAFQ 0 58 5 48 507 0 80
4
2 700
2017
IAFQ 6 69 2 42 538 0 110
4
5 772
Libya
2016
IAFQ 6 79 2 93 24 1 19 0 224
2017
IAFQ 3 84 2 114 25 3 16 0 247
1 Rocket launchers/motors
2 17 stun guns and 1 air gun
3 6 stun guns and 4 air guns
4 Spearguns
Statistical annex
TABLE 2
Arms seized by type, 2010-2017
108
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Morocco
2016
IAFQ 0 46 9 5 65 0 19 0 144
2017
IAFQ 0 37 9 0 102 0 14 0 162
Sudan
2016
IAFQ 7 623 0 341 27 0 0 0 998
2017
IAFQ 16 724 0 573 48 0 0 0 1361
Tunisia
2016 IAFQ 207 30 0 11 687 0 0 0 935
2017
IAFQ 14 16 0 2 601 2 0 0 635
Southern Africa
Botswana
2016 IAFQ 0 3 0 9 14 0 2
5
028
2017
IAFQ 0 12 2 22 23 0 2
5
061
Western Africa
Burkina Faso
2011 SOF 174
2012 SOF 852
2013 SOF 1231
2016
IAFQ 8 41 0 30 270 2 0 7 358
2017
IAFQ 0 0 0 35 183 0 0 0 218
Cabo Verde
2016 UNODA 243
2017 UNODA 144
Cote d'Ivoire
2016 IAFQ 0 31 4 0 5 0 0 151 191
2017
IAFQ 0 59 8 23 32 0 0 115 237
Ghana
2010 SOF 0 7 0 0 5 0 12 0 24
2011 SOF 0 10 0 0 4 0 14 0 28
2012 SOF 0 1 1 10 41 0 20 0 73
5 Pellet guns
109
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
2013 SOF 0 26 4 5 21 0 372 0 428
2016
IAFQ 0 15 0 0 0 0 0 0 15
Guinea
2017 UNODA 57
Niger
2012 SOF 213
Togo
2012 SOF 205
2016
IAFQ 0 0 0 0 83 0 0 0 83
2017
IAFQ 0 0 0 0 83 0 0 0 83
Americas
Caribbean
Antigua and Barbuda
2016 IAFQ 0 12 3 1 0 0 2
6
018
2017
IAFQ 0 4 3 0 1 0 2
7
010
Bahamas
2016 IAFQ 375
2017
IAFQ 350
Cuba
2016 IAFQ 0 4 1 0 0 0 0 0 5
2017
IAFQ 0 1 1 0 0 0 0 0 2
Dominican Republic
2010 SOF 803 141 80 72 1104
2011 SOF 0 732 134 0 66 0 16 0 948
2012 SOF 0 911 250 0 98 0 7 0 1266
2013 SOF 0 501 136 0 70 0 6 0 713
2016
IAFQ 2 1150 225 1 319 1 0 0 1698
2017
IAFQ 0 605 131 0 65 0 0 0 801
Grenada
2016 IAFQ 0 8 0 3 0 1 0 0 12
2017
IAFQ 0 4 0 1 0 0 0 0 5
6 1 converted flare gun and 1 air soft gun
7 2 converted flare guns
110
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Jamaica
2016 IAFQ 0 402 113 37 22 20 55 0 649
2017
IAFQ 0 564 120 67 35 20 56 0 862
Saint Lucia
2016 IAFQ 0 17 2 2 3 0 0 0 24
2017
IAFQ 0 31 2 2 3 0 2 6 46
Trinidad and Tobago
2010 SOF 0 163 102 4 51 7 57 0 384
2011 SOF 151 125 31 53 7 55 425
2012 SOF 186 110 25 51 7 49 429
2013 SOF 194 98 14 37 11 61 419
Central America
Costa Rica
2016 IAFQ 0 3539 2723 300 311 0 4
8
5820 12697
2017
IAFQ 0 3489 2254 265 254 0 7
8
7032 13301
El Salvador
2010 SOF 2182 1558 291 559 20 186 4837
2011 SOF 32 2026 1448 262 473 187 4456
2012 SOF 2 4004 2011 437 878 36 307 7726
2016
IAFQ 0 844 485 155 166 6 0 96 1752
2017
IAFQ 0 1548 795 290 313 21 0 136 3103
Guatemala
2010 SOF 23 2737 805 158 433 416 4580
2011 SOF 4 2375 703 137 452 396 4069
2012 SOF 1 2695 766 74 419 423 4378
2013 SOF 2 2914 744 56 395 447 4558
2016
IAFQ 0 2935 809 96 470 33 597 0 4940
2017
IAFQ 0 2764 743 83 429 39 628 0 4686
Honduras
2016 IAFQ 21 2218 1133 371 274 0 203 0 4220
2017
IAFQ 31 2397 1197 443 315 0 225 0 4608
8 Grenade launchers
111
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Mexico
2010 SOF 100 9664 2961 13826 377 3363 30291
2011 SOF 77 9966 3230 17191 5035 367 5130 40996
2012 SOF 75 6305 1853 11346 2406 255 3563 25803
2013 SOF 23 2236 506 3749 542 103 0 1600 8759
2016
IAFQ 23 2281 401 2030 261 60 375 892 6323
2017
IAFQ 23 2304 379 2834 317 69 423 892 7241
Panama
2010 SOF 44 662 753 234 243 0 0 0 1936
2011 SOF 29 1143 1360 192 240 0 47 0 3011
2012 SOF 21 795 733 162 186 0 25 33 1955
2013 SOF 11 846 611 111 137 0 25 29 1770
2016
IAFQ 0 544 431 103 85 15 11 54 1243
2017
IAFQ 0 580 436 127 137 14 11 46 1351
Northern America
Canada
2012 CND 32000
2016 CND 25123
United States of America
2013 ATF 1382
9
128083
10
45795 40221 28898 397
11
268 245044
12
2014 ATF 717
9
133759
13
43799 38854 26970 361
11
245 244705
12
2015 ATF 765
9
152498
14
44792 41273 27519 581
11
1918 269346
12
2016 ATF 876
9
174836
15
44855 40449 26328 351
11
465 288160
12
2017 ATF 1265
9
201595
16
45407 44085 27241 321
11
347 320261
12
9 May include submachine guns
10 Includes 2195 derringers
11 Includes destructive devices, combination guns and flare guns, tear gas launchers
12 Includes firearms submitted for tracing to the Bureau of Alcohol, Tobacco, Firearms and Explosives by a law enforcement agency, resulting from seizure as well as abandonment, buy-back program, or other recovery method.
Moreover, only seized firearms submitted for tracing are included. Firearms submitted for tracing after recovery do not represent the entire set of all seized firearms. The classification encompasses destructive devices, tear gas
launchers.
13
Includes 2197 derringers
14 Includes 2234 derringers
15 Includes 2094 derringers
16 Includes 2087 derringers
112
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
South America
Argentina
2010 SOF 1 456 787 296 331 1 0 0 1872
2011 SOF 0 704 1288 249 258 3 0 0 2502
2012 SOF 1 2671 4845 1018 1021 9 3 0 9568
2013 SOF 0 1517 1519 696 481 2 3 0 4218
2014 SOF 0 580 549 206 165 1 0 0 1501
2016
IAFQ 334 4941 13282 2054 2723 24 306 0 23664
2017
IAFQ 21 3218 10888 1840 3137 6 547 0 19657
Bolivia (Plurinational State of)
2016 IAFQ 2 21 36 10 53 0 0 0 122
2017
IAFQ 4 13 21 42 72 0 0 0 152
Brazil
2013 SOF 45 3073 7888 413 1590 31 8 0 13048
2016
IAFQ 9 2616 1452 550 134 27 0 6 4794
2017
IAFQ 4 1510 189 472 60 42 0 7 2284
Chile
2010 SOF 0 506 1106 40 1327 0 401 80 3460
2011 SOF 4 669 1113 43 1377 0 493 0 3699
2012 SOF 4 589 1068 45 1313 0 612 0 3631
2013 SOF 6 887 1855 172 1502 0 592 1215 6229
2016 UNODA 7833
2017 UNODA 4137
Colombia
2010 PP 18 10516 23625 1289 8915 111 90 0 44564
2011 PP 19 8235 20171 819 8524 99 96 0 37963
2012 PP 34 7956 18812 2070 9476 158 393 0 38899
2013 PP 21 6934 16462 568 9941 101 113 0 34140
2014 PP 13 6337 14362 493 10111 94 33 0 31443
Colombia
2015 PP 5 6160 12529 444 9795 56 36 0 29025
2016 PP 11 5226 11058 332 8970 52 28 0 25677
113
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
2017 PP 19 4992 9479 467 8698 88 254 0 23997
Ecuador
2010 SOF 5 44 385 3 52 11 514
2011 SOF 7 203 3698 6 306 324 4606
2012 SOF 6 432 3053 17 604 826 5171
2013 SOF 9 310 2323 25 636 218 3739
2016
IAFQ 0 304 1945 16 1017 0 38 3320
2017
IAFQ 1 102 381 1 336 26 847
Guyana
2016 IAFQ 0 47 34 1 28 0 0 0 110
2017
IAFQ 0 68 21 12 18 0 0 0 119
Paraguay
2016 IAFQ 0 36 1 0 0 0 0 0 37
Peru
2010 SOF 0 681 1017 145 358 0 1571 48 3820
2011 SOF 0 637 880 81 148 0 773 183 2702
2012 SOF 0 624 616 49 92 0 9 26 1416
2013 SOF 0 777 1107 101 236 0 101 30 2352
2016
IAFQ 14 793 551 29 59 0 182 0 1628
2017
IAFQ 13 2129 1914 32 481 0 217 0 4786
Suriname
2016 IAFQ 0 57 13 3 95 0 19 0 187
2017
IAFQ 0 52 19 3 81 0 38 0 193
Uruguay
2013 SOF 0 557 1211 398 324 4 0 146 2640
2016
IAFQ 3891
2017
IAFQ 3997
Asia
Central Asia
Kazakhstan
2016 IAFQ 0 23 5 531 4547 0 1526 3 6635
2017
IAFQ 0 24 5 427 3612 0 633 3 4704
114
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Kyrgyzstan
2017 IAFQ 1 31 0 0 43 0 0 5 80
Tajikistan
2016 IAFQ 0 5 0 9 16 0 2 0 32
2017
IAFQ 0 6 0 5 8 0 9 0 28
Eastern Asia
Japan
2016 IAFQ 0 0 341 14 66 0 268
17
0 689
2017
IAFQ 0 0 361 22 60 0 427
17
0 870
South-Eastern Asia
Myanmar
2016 IAFQ 0 67 15 88 36 0 0 0 206
2017
IAFQ 0 79 12 86 46 0 0 0
223
Philippines
2016 IAFQ 0 10 3 3 5 0 0 0 21
2017
IAFQ 0 19 12 2 0 5 0 0 38
Southern Asia
Nepal
2016 IAFQ 0 15 1 4 0 1 2
18
730
2017
IAFQ 0 10 0 16 0 0 2
18
331
Western Asia
Azerbaijan
2016 IAFQ 0 97 7 1 52 2 60 0 219
2017
IAFQ 0 65 4 2 28 0 113 0 212
Iraq
2010 SOF 467 1444 0 9016 22 0 0 0 10949
2011 SOF 605 1443 0 15158 75 0 0 0 17281
2012 SOF 677 1459 0 9624 60 0 0 0 11820
Kuwait
2013 SOF 0 61 0 12 11 0 0 0 84
17 Includes air guns and machine guns
18 Air guns
115
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Lebanon
2016 IAFQ 5 106 4 245 980 3 13 3 1359
2017
IAFQ 1 656 5 188 676 1 20 1 1548
Qatar
2016 IAFQ 7 19 2 0 28 0 0 15 71
2017
IAFQ 4 31 1 0 15 0 0 27 78
Turkey
2013 SOF 3 190 417 7587 1604 0 52 0 9853
Europe
Eastern Europe
Belarus
2016 IAFQ 7416
2017
IAFQ 9066
Hungary
2016 IAFQ 66 83 12 79 11 83 1
19
2 337
2017
IAFQ 0 18 0 93 7 0 64
19
11 193
Poland
2010 SOF 168 262 263 167 168 37 118 1086 2269
2011 SOF 158 201 201 159 158 47 84 569 1577
2012 SOF 184 216 215 184 184 17 102 748 1850
2013 SOF 162 219 220 161 161 23 86 535 1567
Republic of Moldova
2016 IAFQ 0 10 1 6 9 0 6 5 37
2017
IAFQ 0 15 0 3 3 0 6 4 31
Romania
2010 SOF 1 2 5 11 99
2011 SOF 3 1 2 5 84
2012 SOF 2 2 86
2013 SOF 6 2 86
2016
IAFQ 0 2 2 4 3 0 17 0 28
2017
IAFQ 0 4 2 7 0 0 44 0 57
19 Air rifle(s)
116
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
Russian Federation
2010 SOF 47
2011 SOF 2233
2012 SOF 18
2013 SOF 56
2016
IAFQ 0 0 0 0 0 0 0 17197 17197
2017
IAFQ 0 0 0 0 0 0 0 13855 13855
Slovakia
2016 IAFQ 0 842 89 707 41 0 0 36 1715
2017
IAFQ 0 219 60 186 40 0 0 5 510
Ukraine
2016 IAFQ 70 0 740 42 42 30 389 1725 3038
2017
IAFQ 123 0 1016 109 109 23 737 1645 3762
Northern Europe
Denmark
2017 IAFQ 5 230 83 121 175 46 113 5 778
2017
IAFQ 3 284 75 100 194 42 152 4 854
Finland
2010 SOF 39 386 386 677 677 30 2783
2011 SOF 43 448 449 686 770 22 3023
2012 SOF 55 399 399 686 570 18 2127
2013 SOF 120 413 414 662 612 29 3149
Latvia
2010 SOF 3 8 8 1 2 0 1 0 23
2011 SOF 10 0 81 23 4 0 2 0 120
2012 SOF 0 0 15 8 4 0 0 0 27
2013 SOF 0 2 8 2 2 0 2 0 16
Lithuania
2010 SOF 2 0 0 34 74 0 62 11 183
2011 SOF 14 0 0 32 47 1 70 6 170
2012 SOF 3 4 0 16 46 0 39 7 115
2013 SOF 7 7 2 30 72 1 68 11 198
117
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
2016 IAFQ 0 150 81 4 68 0 1
20
5 309
2017
IAFQ 4 119 87 0 51 0 1
20
0 262
Norway
2016 IAFQ 4 127 42 70 25 38 244
21
0 550
2017
IAFQ 3 82 70 38 35 13 75
21
0 316
Sweden
2016 IAFQ 39 430 106 92 158 30 102 0 957
2017
IAFQ 42 401 77 76 106 29 81 0 812
United Kingdom
2016 PP 0 317 223 27
22
223 23 132 1 946
2017 PP 0 274 203 27
22
249 34 160 0 947
Southern Europe
Albania
2016 IAFQ 4 169 5 273 80 0 0 0 531
2017
IAFQ 0 139 5 189 99 0 0 0 432
Croatia
2016 IAFQ 97 655 99 966 0 60 0 1 1878
2017
IAFQ 78 568 75 937 0 26 0 2 1686
Greece
2016 IAFQ 2 489 175 92 7838 57 492 0 9145
2017
IAFQ 2 367 156 163 1504 52 247 1 2492
Italy
2016 IAFQ 6 1296 383 0 4631 3 0 298 6617
2017
IAFQ 0 1045 323 4 2768 9 0 396 4545
Montenegro
2010 SOF 21 195 11 22 81 4 15 379
2011 SOF 19 273 10 24 74 3 24 453
2012 SOF 11 145 14 30 79 9 3 319
2013 SOF 22 162 12 53 110 4 15 411
20 Grenade launcher
21 Blank firing guns and air guns
22 Includes 8 assault rifles
118
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
2016 UNODA 482
2017 UNODA 377
North Macedonia
2010 SOF 9 101 10 87 0 4 6 1062 1279
2011 SOF 1 81 10 49 0 136 15 22 314
2012 SOF 65 5 68 14 7 179
2016
IAFQ 0 46 5 14 38 1 30 0 134
2017
IAFQ 0 36 1 9 35 0 109 0 190
Portugal
2016 IAFQ 0 298 150 30 453 13 2395
23
3415 6754
2017
IAFQ 0 199 95 31 373 7 2208
23
3012 5925
Serbia
2011 SOF 0 223 222 190 0 0 33 0 668
2012 SOF 1258
2016
IAFQ 11 605 37 410 0 11 193 0 1267
2017
IAFQ 25 610 41 1041 0 9 155 0 1881
Slovenia
2016 IAFQ 0 111 0 133 0 0 54 304
2017
IAFQ 0 116 0 149 0 0 59 0
324
Spain
2010 PP 0 801 261 990 3667 4 5723
2011 PP 4 878 315 1061 3731 24 6013
2012 PP 2 2249 1022 1302 4595 208 9378
2013 PP 0 1590 692 1268 3951 14 7515
2014 PP 8 1480 561 1305 3525 18 6897
2015 PP 2 1262 507 1177 3177 36 6161
2016
IAFQ 1 1549 1216 1563
24
4657 17 0 0 9003
23 Includes alarm weapons, sticks, electric weapons, gas, razors, throwing stars and "beasts".
24 Includes 29 assault rifles
119
Statistical annex
Region Year Source
Machine
gun
Pistol Revolver Rifle Shotgun
Submachine
gun
Other Unknown Total
2017 IAFQ 3 4389 1185 1597
25
3303 125 0 0 10602
Western Europe
Belgium
2016 IAFQ 13 687 429 2086 0 13 0 517 3745
2017
IAFQ 19 747 381 1903 0 10 0 554 3614
France
2013 SOF 14 222 60 0 259 7 109 154 825
Liechtenstein
2016 IAFQ 32
2017
IAFQ 13
Luxembourg
2016 IAFQ 1 39 31 52 51 15 547 30 766
2017
IAFQ 0 27 27 41 16 4 748 0 863
Netherlands
2010 IAFQ NND 1405
2016
IAFQ 163
26
1619
26
522
26
753
26
06
26
2336
26
0 5399
26
2017 IAFQ 132
26
1651
26
555
26
991
26
024
26
2110
26
0 5463
26
Oceania
Australia and New Zealand
Australia
2016 IAFQ 17 335 13234 3742 0 742 6103 25502
2017
IAFQ 13 357 14806 4177 0 1145 5874 27819
25 Includes 47 assault rifles
26 Includes found and surrendered arms
120
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
TABLE 3
Arms seized by legal justification of seizures, 2016-2017
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
Africa
Eastern Africa
Burundi
2016 193 193
2017 42 42
Kenya
2016 64 424 4762 14 5264
2017 9.728 9728
Middle Africa
Angola
2016 6056 5.419 11475
2017 4965 5.23 10195
Central African Republic
2016 74 2 4 224 304
2017 1 104 7 4 14 81 211
Northern Africa
Algeria
2016 101 322 87 0 190 700
2017 164 293 65 0 250 772
Libya
2016 224 224
2017 247 247
Tunisia
2016 13 110 392 160 260 935
2017 17 110 358 11 139 635
Western Africa
Cote d'Ivoire
2016 180 11 191
2017 190 22 25 237
121
Statistical annex
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
Togo
2016 83 83
2017 83 83
Americas
Caribbean
Antigua and Barbuda
2016 0 0 18 0 0 0 18
2017 0 0 10 0 0 0 10
Cuba
2016 5 5
2017 2 2
Dominican Republic
2016 0 0 1498 0 135 65 1698
2017 0 0 653 0 125 23 801
Grenada
2016 12 12
2017 5 5
Jamaica
2016 55 48 277 269 649
2017 56 50 392 364 862
Saint Lucia
2016 24 24
2017 26 20 46
Central America
Costa Rica
2017 2
El Salvador
2016 96 1351 1447
2017 132 1898 2037
122
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
Guatemala
2016 1 0 3223 0 404 1312 0 4940
2017 0 0 3200 0 315 1171 0 4686
Honduras
2016 0 210 2612 0 1237 161 4220
2017 0 136 2844 0 1431 197 4608
South America
Argentina
2016 23664 23664
2017 19657 19657
Bolivia (Plurinational State of)
2016 122 122
2017 83 69 152
Brazil
2016 0 0 4377 17 0 100 300 4794
2017 0 0 1769 204 0 61 250 2284
Ecuador
2016 339 2981 3320
2017 303 544 847
Peru
2016 0000591.569 0 1628
2017 0 0 0 15 1118 3.653 0 4786
Asia
Eastern Asia
Japan
2016 624 4 22 39 689
2017 794 29 47 870
South-Eastern Asia
Myanmar
2016 206 206
123
Statistical annex
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
2017 223 223
Philippines
2016 11 5 2 3 21
2017 33 1 4 38
Southern Asia
Nepal
2016 30 30
2017 28 3 31
Western Asia
Azerbaijan
2016 0 4 150 9 56 0 0 219
2017 0 7 152 4 49 0 0 212
Qatar
2016 56 56
2017 49 2 51
Europe
Eastern Europe
Belarus
2016 7416 7416
2017 9066 9066
Hungary
2016 210 127 337
2017 193 193
Republic of Moldova
2016 2 29 3 2 1 37
2017 3 19114331
Romania
2016 0 0 28000028
2017 0 0 57000057
124
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
Northern Europe
Denmark
2016 512 106 160 778
2017 600 78 176 854
Lithuania
2016 289 8 12 309
2017 233 3 19 7 262
Sweden
2016 149 3
2017 135 5
United Kingdom
2016 0 0 856 23 63 0 4 946
2017 0 0 903 7 37 0 0 947
Southern Europe
Albania
2016 435 11 5 80 531
2017 326 6 1 99 432
Greece
2016 6435 2710 9145
2017 2 2490 2492
Portugal
2016 19 3012 114 1338 4498
2017 6 64 2481 152 780 3483
Slovenia
2016 0 0 304 0 304
2017 0 0 324 0 324
Spain
2016 4 568 936 301 336 6858 9003
2017 429 98 332 349 9394 10602
125
Statistical annex
Year
Altered
markings
Illicit
manufacture
Illicit
possession
Illicit
trafficking
Illicit
use
Other Unknown Total
Western Europe
Liechtenstein
2016 32 32
2017 13 13
Luxembourg
2016 0 0 445 0 321 0 0 766
2017 0 0 512 0 351 0 0 863
Netherlands
2016 0 0 3956 64 760 477 142 5399
2017 0 0 4282 43 517 439 182 5202
Oceania
Australia and New Zealand
Australia
2016 18 6 3018 15 33 15621 6791 25502
2017 24 12 2869 26 105 16749 8034 27819
126
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
TABLE 4
Parts and components and ammunition seized, 2016-2017
Region Year
Parts and
components
Rounds of
ammunition
Africa
Eastern Africa
Kenya
2016 0 21141
2017 0 33608
Middle Africa
Angola
2016 1.259 17377
2017 489 66077
Central African Republic
2016 8724
2017 23289
Northern Africa
Algeria
2016 44 56321
2017 118 145668
Libya
2016 47 4299
2017 56 173
Morocco
2016 7 5472
2017 8 4724
Tunisia
2016 28 17959
2017 6 11241
Americas
Caribbean
Cuba
2016 7426
2017 7956
Grenada
2016 1 239
2017 0 32
Central America
Costa Rica
2016 0
2017 0
Guatemala
2016 0 40169
2017 0 44415
Northern America
United States of America
2016 1063
2017 1817
127
Statistical annex
Region Year
Parts and
components
Rounds of
ammunition
South America
Argentina
2016 68 742
2017 204 610
Brazil
2016 629 293211
2017 2690 174848
Ecuador
2016 7809 29472
2017 660 13480
Guyana
2016 630 1516
2017 1.685 1751
Peru
2016 64 47917
2017 202 145061
Asia
Eastern Asia
Japan
2016 1 14724
2017 3 21048
South-Eastern Asia
Myanmar
2016 157 29493
2017 171 16748
Philippines
2016 43 96
2017 82 146
Southern Asia
Nepal
2016 6 28
2017 19 62
Western Asia
Azerbaijan
2016 78 5851
2017 52 4385
Qatar
2016 12 7209
2017 3 7501
Europe
Eastern Europe
Romania
2016 0 81861
2017 0 79810
128
2020
GLOBAL STUDY ON FIREARMS TRAFFICKING
Region Year
Parts and
components
Rounds of
ammunition
Northern Europe
Lithuania
2016 67 6893
2017 54 5911
Sweden
2016 20 3563
2017 16 3031
United Kingdom
2016 5
2017 14
Southern Europe
Croatia
2016 163 636
2017 104 695
Greece
2016 1823 578912
2017 862 136241
North Macedonia
2016 6 4995
2017 1 9718
Slovenia
2016 58 48214
2017 515 25151
Spain
2016 2270 8813
2017 118 10191
Oceania
Australia and New Zealand
Australia
2016 2509 309980
2017 5369 5196801
129
Statistical annex
TABLE 6
Data sources
TABLE 5
SDG Indicator 16.4.2 (seized arms only) of selected countries for the period, 2016-2017
Country A B C
D
(A+B+C)
EFGH
I
(D+E+F+G+H)
SDG
(D/I,
percentage)
Antigua and Barbuda 0 2 0 2 0 14 9 25 8
Argentina 5842 0 0 5842 15040 11022 0 0 31904 18
Australia 357 0 150 507 2975 3 28881 20530 52896 1
Azerbaijan 224 0 0 224 0000 224 100
Bahamas 91 493 584 0 2 586 100
Brazil 83 684 75 842 0 5290 0 6132 14
Grenada 0 11 11 1 5 17 65
Kenya 0 0 0 9722 6 9728 0
Lithuania 3 3 176 179 2
Peru 2875 0 0 2875 788 0 0 3663 78
Republic of Moldova 5 5 48 53 9
Romania 0 0 0 0 0 19 61 0 80 0
Spain 401 737 1138 25 898 2061 55
United Kingdom 94 145 18 257 0 114 1370 0 1741 15
A Weapon seized from illegitimate owner and weapon found in national registry (e.g., lost or stolen) (national tracing)
B Point of diversion of the weapon (last legal record) identified through tracing and weapon found in foreign registry (international tracing)
C Point of diversion otherwise established by a competent authority
D SDG numerator: successfully traced arms of illicit origin
E Tracing attempted, but not enough information to identify point of diversion
F Tracing procedure still pending
G No tracing procedure initiated
H Unknown status with respect to marking
I SDG denominator: Potentially traceable arms of illicit origin
SDG SDG Indicator
Data source
IAFQ
UNODC Illicit Arms Flows Questionnaire
PP
Official communication
SOF
UNODC Study on Firearms 2015
WCO
World Customs Organization
UNODA
United Nations Office for Disarmament Affairs
CND
Commissioner of Firearms report and Statistics Canada catalogue
NND
Nationaal dreigingsbeeld 2012
ATF
Bureau of Alcohol, Tobacco, Firearms and Explosives
UNPOL
United Nations Police
NA
Not available / Not applicable
