Content uploaded by Nils Duquet
Author content
All content in this area was uploaded by Nils Duquet on Aug 27, 2015
Content may be subject to copyright.
Report
June 2015
Firearms and Violent Deaths
in Europe
Nils Duquet & Maarten Van Alstein
Firearms and Violent Deaths in Europe
An Exploratory Analysis of the Linkages Between Gun
Ownership, Firearms Legislation and Violent Death
FIREARMS AND VIOLENT DEATHS IN EUROPE P 1
Contents
INTRODUCTION __________________________________________________________________ 2
1 FIREARMS IN EUROPEAN SOCIETIES __________________________________________ 5
1.1 Lack of reliable and comparable administrative data on gun ownership __________________________ 5
1.2 Comparing the results of self-reporting surveys on gun ownership ______________________________ 6
1.3 Expert estimates _____________________________________________________________________ 16
2 GUN-RELATED DEATHS IN EUROPE _________________________________________ 19
2.1 Suicide _____________________________________________________________________________ 21
2.2 Homicide ___________________________________________________________________________ 25
3 GUN OWNERSHIP AND VIOLENT DEATH: LESSONS FROM PREVIOUS RESEARCH ______ 32
3.1 Individual-level versus population-level studies ____________________________________________ 32
3.2 Suicide _____________________________________________________________________________ 33
3.3 Homicide ___________________________________________________________________________ 38
4 GUN OWNERSHIP AND VIOLENT DEATHS IN EUROPE: A NEW ANALYSIS ____________ 41
5 THE IMPACT OF STRICTER GUN LEGISLATION? ________________________________ 46
5.1 The impact of the 1997 firearms legislation in Austria _______________________________________ 49
5.2 The impact of the 2006 weapons act in Belgium ____________________________________________ 51
5.3 The impact of military service reforms in Switzerland in 2003 _________________________________ 54
CONCLUSION ___________________________________________________________________ 56
ANNEX _____________________________________________________________________ 61
ENDNOTES ____________________________________________________________________ 74
FIREARMS AND VIOLENT DEATHS IN EUROPE P 2
Introduction
On a regular basis, news stories appear in the media about public shootings where shooters use their
guns to open fire and kill people in shopping malls or on school campuses. Mostly these stories deal with
incidents in the United States. Over the last years, however, a number of European countries have
experienced similar public shooting incidents. Notable cases were the shootings at Tuusula and
Kauhajoki in Finland (2007 and 2008), the killings in Cumbria in the UK (2010), the Utøya attacks by
Anders Breivik in Norway (2011), and the shootings at Alphen aan den Rijn in the Netherlands and Liège
in Belgium in 2011. Public shootings draw a high level of media attention. Less striking in the public eye,
but not less significant – not least in quantitative terms –, are the numbers of people in Europe killed by
firearms in the context of gun-related crime or in domestic shootings. It is estimated that between 2000
and 2010, over 10,000 victims of murder or manslaughter were killed by firearms in the 28 Member
States of the European Union (EU). Every year, over 4000 suicides by firearm are registered in the EU.
This means that, on average, there are 0.24 homicides and 0.9 suicides by firearm per 100,000
population in Europe every year.
1
Compared with the US or other countries around the globe, the rates of gun-related violent death in
Europe are rather low, certainly where the homicide rates are concerned. This does not mean, however,
that the problem of gun violence has not appeared on the European policy radar in recent years. On the
contrary, the attention devoted to the problem by law enforcement agencies and policy-makers has
been growing. Reacting not only to shooting incidents such as those mentioned above, but also to
warnings by police and law enforcement agencies that criminals are increasingly willing to use (heavy)
firearms and that illegal trafficking in firearms is on the rise, a number of European countries have
announced policy interventions targeted at reducing levels of gun-related violence and crime. The
European Commission has also become an active actor in firearms policy. In October 2013 it announced
a plan to reduce gun violence in Europe, in which it defined the misuse of firearms, whether legally-
owned or illicitly manufactured or acquired, as “a serious threat to the EU’s security from both an
internal and an external perspective”. One of the major problems the Commission identified in its initial
policy papers was the problem of a lack of sound and adequate knowledge about firearms in Europe.
The commission noted that “a lack of solid EU-wide statistics and intelligence hampers effective policy
and operational responses”. One of the ambitions of the EU’s firearms policy is, therefore, to address
the gaps in knowledge concerning gun violence.
2
An additional problem is that the lack of reliable and comprehensive information on firearms in Europe
is not limited to the sphere of law enforcement and policy-making. European scholarly research focusing
specifically on firearms availability, gun control and gun-related violence is scarce. There is a research
FIREARMS AND VIOLENT DEATHS IN EUROPE P 3
community in Europe focusing on small arms and light weapons (SALW), but it is predominantly
concerned with the export of firearms and the connections between these arms flows and violence in
developing, transitional or fragile states outside Europe. Scientific research on firearms and gun-related
violence in the domestic European context is much less advanced. The scanty research efforts made in
this field by epidemiologists, criminologists and legal scholars remain fragmented, and suffer from the
fact that there is no integrated scholarly community dealing with gun-related issues. Language barriers,
moreover, often prevent the wider dissemination of research results. Given this relative lack of
European firearms research, American studies are still clearly dominant at present in research on the
links between the availability of firearms and gun-related violence. Greene and Marsh have calculated
that out of the 665 studies on firearms and violence that they reviewed, 64% were about the USA. Of
the remaining studies not on the USA, 13% concerned cross-national comparisons or articles in which
the geographical focus was unspecified (such as reviews), while 8% were about developing countries.
Only 15% concerned other developed countries such as Canada, Australia, the UK and Germany.
3
Given
the particularities of the American context, and more specifically the fact that the US has one of the
highest rates of gun-related deaths and crime among industrialized democracies, simply transposing the
results of American research to the European context is problematic.
What are the levels of firearms availability in Europe? Are there links between the levels of gun
ownership in European countries and these countries’ rates of violence and violent death? And what is
the impact of European gun laws on public safety and health? The absence of evidence specifically for
the European context makes it difficult for policy-makers and researchers to find impartial and unbiased
answers to these questions. Hence the pressing need for research that specifically focuses on gun-
related violence in the European context: and with the present report, we would like to make a
contribution to that effort. As we are moving into largely uncharted territory, our analysis of the
European situation will necessarily be exploratory. Our primary ambition is to collect and take stock of
the fragmented evidence that is available on gun-related violence in Europe. Our geographical coverage
will be broader than the EU and encompasses a group of approximately 40 European countries,
although in some instances we will limit our analyses to the EU28.
In the report’s first chapter, we briefly dwell on one of the most crucial variables in research on gun
control and violence: the level of gun ownership in society. Although the prevalence or availability of
firearms is a key variable, collecting adequate data on levels of gun ownership can be troublesome. In
chapter 1 we therefore devote some space to a critical assessment of the available statistics for Europe.
Next, in chapter 2, we look at gun-related violence in Europe. Given the absence of good data on gun-
related violence in general, including information not only on mortality but also on injuries and other
forms of firearms-related victimization, we will focus exclusively on violent deaths – which seems a
legitimate methodological choice for exploratory purposes. We urge the reader, however, to keep in
mind that gun-related violence is a much more complex phenomenon than this focus might suggest. As
FIREARMS AND VIOLENT DEATHS IN EUROPE P 4
is normal in research dealing with gun control not only from a public safety but also a public health
perspective
4
, we shall look both at gun-related homicides and at suicides. Taking the analysis further, we
then ask in chapters 3 and 4 whether rates of gun possession and violent death in Europe are correlated:
do high levels of gun possession in European countries correlate with high levels of homicide and
suicide? The results of probing that question lead us to suggest that research into gun possession and
violent death should also factor in the effects of firearms legislation. Specific European research into this
question is scarce, which makes it difficult at the moment to arrive at conclusions for the whole of
Europe. In chapter 5 we therefore focus on the results of three recent studies on the effects of stricter
gun legislation on violent death rates in Austria, Belgium and Switzerland.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 5
1 Firearms in European Societies
The rate of firearms possession in a country is a crucial variable in research on the relationship between
guns and violent death. Finding out how many guns there are in European countries, however, is a
challenging endeavour. Calculating the number of legally-held firearms in Europe is not an easy task
given the lack of reliable and comparable official data on gun possession, but reliable statistics on the
number of illegally-held and trafficked firearms in Europe are even harder to find. In the first place, of
course, this reflects the covert nature of illicit possession, but it is also a function of the lack of reliable
and comparable estimates on illegal gun possession and trade for all European countries. The dearth of
reliable data becomes very plain in a recent study commissioned by the European Commission where
the number of illegally-held firearms is estimated as lying somewhere between 81,000 and 67 million.
5
This clearly illustrates how hard it is to find out how many firearms there are in Europe. In this chapter
we will give an overview of the available figures on European gun ownership rates in Europe, looking
particularly at different sources for data on gun prevalence such as administrative registries, self-report
surveys and proxies, and paying attention to their specific merits and shortcomings.
1.1 Lack of reliable and comparable administrative
data on gun ownership
As already stressed, no reliable and comparable administrative data on gun possession across Europe
are currently available. An important explanation for this is that not all European countries have a
centralized database to register legally-held firearms. As a positive development, Directive 2008/51/EC
(amending the Firearms Directive 91/477/EEC) stipulated that all EU Member States must establish by
31 December 2014, and then maintain, a computerized data filing system in which details of all civilian
possessed firearms must be recorded
I
. At the moment it is unclear to what extent EU Member States
have put in place the required data filing systems. What is clear, however, is that the existence of such
data filing systems in all Member States in the near future will improve the authorities’ chances of
knowing how many and what types of firearms are legally held by civilians in the EU. To achieve this, it
will be vital to develop good procedures for data input to ensure reliable data, and to develop software
that allows for fast and easy data analysis.
Given their illicit character, the number of illegally-held firearms is harder to estimate. In a recent
communication, the European Commission stated that according to the Schengen Information System,
I
Article 4 of the revised Directive states that each firearm that is subject to the Directive must be recorded in this (centralized
or decentralized) data filing system. The filing system must record, and preserve for not less than 20 years, each firearm’s
type, make, model, calibre and serial number, as well as the names and addresses of the supplier and the person acquiring or
possessing the firearm.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 6
almost half a million lost or stolen firearms remain unaccounted for in the EU.
6
To estimate the number
of illegally held guns, a number of studies have looked at official figures - for example, statistics on
seized guns - and used the “capture-recapture” method to estimate the scale of illegally-held firearms in
society. One of the most accurate estimates of illegal firearms possession using this approach has been
made in the Netherlands. Based on the available police data on firearms-related incidents, a 2005 study
using the capture-recapture method suggested the figure of approximately 54,000 persons illegally
possessing firearms in the country in 2001-2003.
7
In most other European countries, however, similarly
reliable estimates are lacking. Given the different characteristics of illicit firearms markets in different
countries, the Dutch figures cannot be extrapolated to the rest of the EU. We must therefore conclude
that it is currently impossible to estimate the total number of legally and illegally held firearms in Europe
based on official statistics. Other data sources and methods are needed to estimate gun ownership rates
in Europe.
1.2 Comparing the results of self-reporting surveys
on gun ownership
Self-reporting through surveys is generally considered a valuable way of generating data on gun
ownership rates, but there are a number of methodological pitfalls and limitations linked to it. First of
all, these surveys only provide us with insight into the level and nature of legal firearms possession. We
can assume that illegal gun owners are generally not inclined to give truthful answers to survey
questions about their firearms possession. The most often cited problem is the potential for invalid
responses as a result of the often controversial nature of (even legal) gun possession in many countries.
8
Having firearms in the house may be perceived by respondents as sensitive behaviour, which may lead
to a social desirability bias and therefore an underestimation of ownership rates. If surveys ask about
the guns present in households, another problem is that respondents may misreport about guns kept by
other household members. More specifically, women tend to under-report guns present in their home,
or may even not be aware that firearms are kept by others in their household.
9
These limitations
notwithstanding, self-reporting offers a useful tool for estimating rates of gun possession. Interestingly,
a number of American studies comparing self-reported personal gun ownership with data from
administrative records have revealed low false-negative rates.
10
Although we must be cautious in
transposing these findings to the European context, where the social desirability bias might work
differently than in the US, these studies underline the value of using this type of data for estimating
firearms possession in society. In the following sub-sections we summarize the results of our own
analyses of data from two major self-reporting studies on firearms possession in Europe: the
International Crime Victims Survey and the Flash Eurobarometer 383: Firearms in the European Union.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 7
International Crime Victims Survey (2004-2005)
The International Crime Victims Survey (ICVS) is a solid longitudinal self-reporting study developed to
monitor and study crime and the perception of crimes, and attitudes towards the criminal justice
system, in a comparative international perspective. Between 1989 and 2005 more than 300,000 people
in 78 different countries were interviewed, using standardized questionnaires, about their experiences
with victimization and related topics.
I
Interestingly, these surveys included questions on gun ownership
in the respondents’ households.
II
With regard to gun ownership, data for 24 European countries are
available covering the years 2004-2005. In this sub-section we will give an overview of the most
important findings of analyses carried out on these data (see figure 1).
Analysis of the ICVS data indicates a strong variation in national firearms ownership rates in Europe. The
countries included in the survey that show the highest household gun ownership rates are Finland
(38%), Switzerland (27%), Norway (26%) and Iceland (24%). Other European countries with rather high
levels of household firearms possession are Greece, Sweden and Portugal. The countries with the lowest
rates in gun ownership are Poland (4%), the Netherlands (5%) and the United Kingdom (6%). Further,
this analysis shows that in most European countries the rate of handgun ownership is considerably
lower than that of general firearms possession. What is particularly interesting is that most of the
countries where a significant proportion of gun-possessing households own a handgun are also the
European countries with the largest firearms production such as Austria, Belgium, Bulgaria, Germany,
Italy and Switzerland. One possible explanation for this observation could be that countries with
significant firearms production have traditionally tended to have a less restrictive legal framework for
civilian handgun ownership.
Besides national figures, the ICVS also published the rates of gun ownership in a number of cities
worldwide, including 22 European cities. Not surprisingly, the highest rates can be found in cities in
the countries with the highest proportions of households owning a firearm (see figure 2).
I
The International Crime Victims Survey (ICVS) is an international victimization survey established to generate international
comparative crime and victimization data. It is a phone survey that is organized on a regular basis. It was conducted for the
first time in 1989 and repeated in 1992, 1996, 2000, 2004-2005 and 2010. For the different rounds of the survey, a total of
more than 300.000 respondents in thirty countries were interviewed about their experiences with crime and victimization.
II
This type of question poses a number of methodological problems (see supra), but the results are nonetheless interesting and
indicative, not least because they allow a comparative insight into gun ownership from an international perspective.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 8
Figure 1: Percentage of households owning a firearm and owning a handgun according to ICVS,
2004-2005
Source: ICVS
11
0.9
1.3
0.4
0.5
0.7
3.6
6.6
2.2
5.2
0.5
7
5.3
4.2
1.2
2.1
1
5.6
3.7
3.9
1.6
1.4
1.4
3.7
10.3
6.3
4.4
4.8
6
6.1
6.7
7
9.7
10.4
11.4
12
12.3
12.4
12.5
12.6
12.7
12.9
15.1
16.1
18.3
19.3
20.6
23.5
26.1
26.6
37.9
0 5 10 15 20 25 30 35 40
Poland
The Netherlands
United Kingkom
England & Wales
Scotland
Estonia
Bulgaria
Hungary
Belgium
Spain
Luxembourg
Ireland
Germany
Denmark
Northern Ireland
Italy
Austria
France
Portugal
Sweden
Greece
Iceland
Norway
Switzerland
Finland
% of houseolds owning a firearm % of households owning a handgun
FIREARMS AND VIOLENT DEATHS IN EUROPE P 9
Figure 2: Percentage of households owning a firearm and owning a handgun in various European
cities, according to ICVS, 2004-2005
Source: ICVS
12
0.7
1.2
0
0.3
0.2
0.1
2.1
4.3
2.5
1.4
3.8
2.9
5.2
5.7
2
1.3
2.1
3.7
2.9
0.6
4.6
6.8
1.5
1.8
1.8
1.8
2
2.2
4.1
5.1
5.4
6.9
7.6
8
8.9
9.9
10
10.1
10.8
12.1
13.7
15.9
16.2
16.2
0 2 4 6 8 10 12 14 16 18
Amsterdam (Netherlands)
Warsaw (Poland)
London (England)
Belfast (Northern Ireland)
Edinburgh (Scotland)
Dublin (Ireland)
Paris (France)
Tallin (Estonia)
Brussels (Belgium)
Copenhagen (Denmark)
Berlin (Germany)
Budapest (Hungary)
Vienna (Austria)
Rome (Italy)
Madrid (Spain)
Stockholm (Sweden)
Athens (Greece)
Lisbon (Portugal)
Oslo (Noray)
Reykjavik
Helsinki (Finland)
Zurich (Switzerland)
% of households owning a firearm % of households owning a handgun
FIREARMS AND VIOLENT DEATHS IN EUROPE P 10
In the ICVS, respondents who stated there was a gun in their household were asked which sort of gun
this was. Our analyses of the ICVS data indicate that in a large number of European countries, the
majority of gun owners possess long guns such as shotguns and rifles. The ownership of handguns
is clearly not equally widespread in Europe. Only in 5 of the 23 European countries covered did
more than 60% of the gun possessors report that they owned a handgun (see table 1).
Table 1: % of types of firearm (excluding air rifles) possessed in European countries, according to ICVS,
2004-2005
Handgun
Shotgun
Rifle
N
Austria
65%
34%
47%
158
Belgium
59%
36%
26%
136
Bulgaria
71%
42%
0%
65
Denmark
18%
58%
56%
123
England & Wales
14%
66%
59%
44
Estonia
52%
50%
13%
120
Finland
31%
73%
58%
326
France
30%
57%
34%
172
Germany
77%
26%
40%
101
Greece
17%
88%
10%
242
Hungary
57%
49%
14%
72
Iceland
7%
82%
49%
392
Ireland
11%
78%
25%
130
Italy
55%
19%
47%
190
Lithuania
57%
47%
10%
68
Luxembourg
70%
30%
50%
74
Netherlands
67%
30%
20%
30
Northern Ireland
22%
76%
20%
112
Norway
13%
64%
44%
1078
Poland
57%
44%
9%
77
Portugal
31%
23%
68%
219
Scotland
17%
63%
48%
52
Spain
11%
81%
7%
177
Sweden
15%
66%
68%
207
Switzerland
39%
0%
76%
993
United Kingdom
13%
67%
57%
46
Source: ICVS
13
, own calculations
FIREARMS AND VIOLENT DEATHS IN EUROPE P 11
Flash Eurobarometer 383: Firearms in the European Union (2013)
A more recent self-reporting study that included a question on gun ownership is the 2013
Eurobarometer poll on firearms policy (Figure 3). This study reported that 5% of the population in the
European Union own a firearm and another 5% used to own a firearm. This would mean that there are
currently at least 25 million gun owners in the EU.
As with our analysis of the older ICVS data, significant differences in national gun possession rates can
be observed. While the share of population owning a firearm is relatively high in Cyprus (18%) and
Finland (13%), it is quite low in countries such as the Netherlands (1%), the United Kingdom (2%), Poland
(2%) and Romania (2%). Follow-up questions about the number of firearms and the type of firearms
owned in the different EU Member States were not included in the Eurobarometer survey.
Some background information on the gun possessors is, however, available. The 2013 Eurobarometer
data indicate, for example, that gun possession is more likely for men than for women in the EU: while
8% of the men said they owned a firearm at the moment and 9% of the men said they had owned a
firearm during their life, these rates were only 1% and 2% among women. In addition, gun ownership in
the EU is more prevalent among older generations and in rural areas. The respondents were also asked
for their reasons for firearms possession. Hunting was the most cited reason for firearms possession in
the EU: 35% of those who owned or used to own a firearm gave hunting as a reason. The two other
main reasons for legal firearms possession in Europe were professional reasons (e.g. police, army,
security services - 29%) and sports (23%). Personal protection and collection were cited by respectively
14% and 5% of the group of current and previous gun owners.
14
FIREARMS AND VIOLENT DEATHS IN EUROPE P 12
Figure 3: Estimates of gun ownership in Europe according to Eurobarometer, 2013
Source: Eurobarometer
15
4
4
5
6
4
6
7
2
4
5
6
7
3
5
7
7
11
3
4
7
7
7
4
6
4
3
13
1
2
2
2
3
3
3
5
5
5
5
5
6
6
6
6
6
7
7
7
7
7
8
8
9
13
18
0 2 4 6 8 10 12 14 16 18 20
Netherlands
United Kingdom
Poland
Romania
Hungary
Estonia
Belgium
Malta
Spain
Luxembourg
Latvia
Germany
Italy
Czech Republic
Portugal
Ireland
Croatia
Greece
Austria
France
Denmark
Bulgaria
Sweden
Lithuania
Slovenia
Finland
Cyprus
% of respondents that own a firearm % of respondents that used to own a firearm
FIREARMS AND VIOLENT DEATHS IN EUROPE P 13
Figure 4: reasons for firearms possession by respondents who currently own or used to own a
firearm, EU28, 2013
Source: Eurobarometer
16
Although in two thirds of the EU Member States hunting was the most cited reason for owning a
firearms, significant national variations with regard to motives can be observed (see table 2). In
countries such as Germany and the Netherlands, for example, sport shootings seemed to be the most
cited reason for gun ownership, while hunting was only cited by small proportion of gun owners as their
reason for gun possession. The observation of country variations in reasons for firearms possession
must, however, be interpreted cautiously since for some countries we only have a small number of
respondents. Even so, an interesting distinction is that in the EU15 countries (i.e., those who were
members before 2004) citizens were apparently more likely to own or to have owned firearms for
hunting and sports, while in the new Member States, gun ownership was more often linked to
professional reasons (especially in Romania, Estonia and Bulgaria) and personal self-defence (especially
in Lithuania, the Czech Republic and Slovakia). Based on the available Eurobarometer figures, we can
also deduce that in almost all EU Member States, the collecting of firearms is a not significant reason for
firearms ownership.
A comparison of the motives given for firearms possession by current and previous owners indicates
that in almost all countries, professional reasons were cited more frequently by previous gun owners
than by current gun owners. The opposite can be observed when looking at sport shooting and
especially hunting: current gun owners cite these reasons much more often than previous gun owners
(see table 12 in Annex). This might suggest an evolution in the reasons for firearms possession over
35%
23%
5%
29%
14%
10%
0
5
10
15
20
25
30
35
40
Hunting Sports Collector Professional
reasons
Personal
protection
Other personal
reasons
FIREARMS AND VIOLENT DEATHS IN EUROPE P 14
time. It is clear that not only the current levels of firearms possession are relevant, but longer-term
trends in ownership rates also need to be taken into account.
Table 2: reasons for firearms possession by respondents currently owning a firearm, per country,
2013
Hunting
Sports
Collector
Professional
reasons
Personal
Protection
Other
personal
reasons
Number of
respondents
Austria
42%
32%
7%
20%
29%
12%
69
Belgium
42%
36%
12%
21%
21%
12%
33
Bulgaria
38%
3%
5%
42%
33%
5%
64
Croatia
37%
6%
2%
30%
17%
13%
54
Cyprus
75%
2%
3%
31%
2%
3%
93
Czech Republic
37%
68%
15%
25%
53%
7%
59
Denmark
73%
17%
5%
3%
3%
7%
77
Estonia
39%
4%
0%
43%
25%
4%
28
Finland
77%
36%
3%
9%
3%
9%
132
France
69%
11%
10%
11%
8%
12%
73
Germany
20%
39%
2%
20%
6%
14%
49
Greece
76%
8%
5%
15%
21%
9%
67
Hungary
42%
13%
0%
7%
39%
16%
31
Ireland
65%
28%
9%
14%
19%
9%
65
Italy
31%
31%
10%
31%
6%
8%
51
Latvia
55%
11%
2%
21%
36%
4%
47
Lithuania
19%
10%
0%
19%
54%
7%
74
Luxembourg
20%
56%
20%
28%
12%
4%
25
Malta
57%
35%
35%
4%
0%
9%
23
Netherlands
10%
60%
0%
20%
10%
0%
10
Poland
44%
33%
0%
28%
22%
0%
18
Portugal
56%
12%
6%
23%
23%
12%
52
Romania
18%
6%
0%
65%
6%
24%
17
Slovakia
34%
26%
11%
37%
51%
3%
35
Slovenia
48%
24%
8%
12%
7%
11%
84
Spain
64%
22%
0%
22%
2%
8%
50
Sweden
73%
28%
8%
3%
0%
9%
80
United Kingdom
44%
44%
20%
19%
19%
6%
16
Source: Eurobarometer, own calculations
International Crime Victims Survey: longitudinal data from 1989-2005
The long series of self-reporting data from the ICVS makes it possible – to a certain extent – to identify
trends in gun ownership rates between 1989 and 2005 for a smaller group of (mainly Western)
European countries (see table 3). The most important observation is that there is no general trend of
increasing or decreasing ownership in European countries. Instead, we observe different trends in
different countries. In Belgium, France, Italy, Norway and Switzerland, the numbers of households
stating that they owned a firearm decreased substantially over the period, while in Finland, Germany,
the Netherlands, Portugal, Sweden, the United Kingdom and Northern Ireland, an opposite evolution
took place and the figures increased. Interestingly, the decrease in gun ownership seems to occur more
frequently in countries with traditional high rates of gun ownership than in the other countries. We can
FIREARMS AND VIOLENT DEATHS IN EUROPE P 15
observe a significant decrease by 2004-2005 in four of the five countries that had the highest household
gun ownership rates in 1989. Unfortunately, a similar long-term data series on gun ownership in Central
and Eastern European countries is lacking.
Table 3: % of households with a gun
I
, 1989-2005
1989
1992-1994
1995-1998
1999-2003
2004-2005
Austria
-
-
15.3
-
15.1
Belgium
16.5
15.6
-
11.4
11.4
Bulgaria
-
-
-
-
9.7
Denmark
-
-
-
11.7
12.6
England & Wales
4.7
4.2
3.9
3.1
6.1
Estonia
-
7.5
8.3
7.4
7.0
Finland
25.5
25.4
28.8
23.7
37.9
France
25.3
-
22.6
18.2
16.1
Germany
9.2
-
-
-
12.5
Greece
-
-
-
-
20.6
Hungary
-
-
-
-
10.4
Iceland
-
-
-
-
23.5
Ireland
-
-
-
-
12.4
Italy
-
16.1
-
-
12.9
Luxembourg
-
-
-
-
12.3
Netherlands
2.0
2.0
2.3
1.7
4.8
Northern Ireland
9.4
-
-
-
12.7
Norway
31.5
-
-
-
26.1
Poland
-
2.5
4.7
3.7
4.4
Portugal
-
-
-
13.2
18.3
Scotland
5.1
-
4.2
3.1
6.7
Spain
12.7
-
-
-
12.0
Sweden
-
15.9
16.2
15.9
19.3
Switzerland
32.8
-
35.3
35.7
28.6
United Kingdom
5.2
-
4.1
3.0
6.0
Source: ICVS
17
I
The question was “Do you or anyone else in your household own a handgun, shotgun, rifle, or air rifle?”
FIREARMS AND VIOLENT DEATHS IN EUROPE P 1 6
Figure 5: % of households with a gun
I
, selected countries, 1989-2005
Source: ICVS
18
1.3 Expert estimates
A third source for calculating the number of guns in Europe is provided by expert estimates. In 2007 the
Small Arms Survey (SAS) published a first detailed assessment of the global distribution of factory-made
civilian firearms. SAS used various sources and methods to estimate the total civilian gun ownership
rates in 178 countries and to provide statistics on the number of weapons per 100 persons.
19
Although
these national estimates by SAS are not equally reliable for every country, they are currently considered
as one of the most reliable sources for international comparisons of gun ownership.
The SAS estimated that the total number of firearms owned by civilians worldwide was approximately
650 million. In table 4 we present the national estimates by SAS for 42 European countries. These
estimates indicate that civilians possessed approximately 93.2 million firearms at the time in all
European societies, and approximately 79.8 million in the EU. This comes down to 15.7 firearms per 100
persons. In absolute figures, a large number of these firearms were thought to be owned in Germany
(25 million) and France (19 million); yet the European countries with the highest percentage rates of
firearms ownership were Switzerland and Finland. In these two countries there were an estimated
number of 45 to 46 firearms for every 100 persons. Other countries with high firearms possession rates
according to the SAS estimate were Serbia, Cyprus, Sweden, Norway, France, Austria, Iceland and
Germany (with more than 30 firearms per 100 persons). At the other end of the spectrum we find
countries such as the Netherlands, Poland and Romania, with estimates of less than 5 civilian firearms
for every 100 persons. These estimates clearly confirm the existence of large national variations in gun
I
The question was “Do you or anyone else in your household own a handgun, shotgun, rifle, or air rifle?”
0
5
10
15
20
25
30
35
40
Belgium France Norway Switzerland Germany the
Netherlands
Northern
Ireland
England &
Wales
1989 1992-1994 1995-1998 1999-2003 2004-2005
FIREARMS AND VIOLENT DEATHS IN EUROPE P 17
possession rates in Europe, even between neighbouring countries. In Germany (30.3) , for example, the
gun possession rate per 100 inhabitants was more than 7 times higher than in the Netherlands (3.9).
Other striking examples are the differences in reported gun ownership rates between the Czech
Republic (16.3) and Slovakia (8.3), or between Denmark (12.0) and Sweden (31.6). In addition, as stated
earlier, it must be borne in mind that large differences in gun possession rates exist even within
countries. Significant differences in gun ownership rates can be observed even within relatively small
countries such as Belgium
20
and Austria
21
.
A comparison of the top-ranked and bottom-ranked countries seem to confirms the estimates of levels
of firearms possession based on other methods. Of course, certain discrepancies exist, some of which
might reflect the different questions posed (individual firearm possession vs household firearm
possession vs estimated numbers of firearms in a society). Some discrepancies, however - especially
relating to Central and Eastern European countries - are striking and cannot be attributed to different
research questions. The SAS estimates for the total number of firearms in Bulgaria, Hungary and Poland,
for example, are lower than the household firearm rates reported in the ICVS data for these countries
(which – for different reasons – can be considered as an absolute minimum for the total number of
firearms in society). These observed differences are presumably not the result of dramatic changes in
gun possession rates in these countries, since the figures are from the same time period, but are very
likely the result of over-conservative estimates. This underlines that the available figures on gun rates
must be used with caution when analyzing the connection between gun ownership and violent death.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 18
Table 4: Estimates of gun ownership in Europe according to the Small Arms Survey (2007)
Estimated number of
civilian firearmsI
Number of firearms
per 100 people
Switzerland
3,400,000
45.7
Finland
2,400,000
45.3
Serbia
3,050,000
37.8
Cyprus
275,000
36.4
Sweden
2,800,000
31.6
Norway
1,400,000
31.3
France
19,000,000
31.2
Austria
2,500,000
30.4
Iceland
90,000
30.3
Germany
25,000,000
30.3
FYR Macedonia
490,000
24.1
Montenegro
150,000
23.1
Greece
2,500,000
22.5
Northern Ireland
380,000
21.9
Croatia
950,000
21.7
Kosovo
415,000
19.5
Latvia
280,000
19.0
Bosnia-Herzegovina
675,000
17.3
Belgium
1,800,000
17.2
Czech republic
1,600,000
16.3
Luxembourg
70,000
15.3
Slovenia
270,000
13.5
Denmark
650,000
12.0
Italy
7,000,000
11.9
Malta
48,000
11.9
Spain
4,500,000
10.4
Estonia
123,000
9.2
Ireland
270,000
8.6
Albania
360,000
8.6
Portugal
900,000
8.5
Slovakia
450,000
8.3
Moldova
300,000
7.1
Ukraine
3,100,000
6.6
Bulgaria
480,000
6.2
England & Wales
3,400,000
6.2
Hungary
560,000
5.5
Scotland
280,000
5.5
Netherlands
510,000
3.9
Poland
510,000
1.3
Lithuania
135,000
0.7
Romania
160,000
0.7
-..7
Total
93,231,000
15.7
Total – EU28
79,801,000
15.7
Source: Small Arms Survey
22
I
The Small Arms Survey presented for each country a “low total estimate”, “a high total estimate” and an “averaged total
estimate”. In this table we give the results of the “averaged total estimate”.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 19
2 Gun-related Deaths in Europe
After looking at the prevalence of guns in European societies, in this section we turn to the number of
violent deaths in Europe and more particularly, gun-related homicides and suicides. Relatively good
statistical information on violent deaths is available for most European countries, although data are
often partial and time series incomplete. This explains why the figures mentioned in policy notes are
often estimates. The best source for international mortality data at the moment is provided by the
World Health Organization (WHO). For data on European countries we can specifically turn to the
European Detailed Mortality Database of the WHO Regional Office for Europe
23
, which contains detailed
mortality data by cause of death. These data are submitted to the WHO by the European member
states. It is important, however, to note some of the problems associated with data from medical
sources. First, the context of death is not revised after a police investigation or trial. This implies that
deaths initially recorded as homicide could have been subsequently determined to be accidental. The
numbers of such discrepancies might be low, but as many European countries have small numbers of
gun-related homicides and accidents, a small number of revisions could have a significant influence on
apparent trends. Further, it is always possible that the medical personnel responsible for generating the
data will misreport the context of a death. They might, for instance, register cases of suicide as
accidents. Another problem is that police and WHO data on homicides can be systematically
mismatched. In Western Europe, WHO data on homicide are for example lower than those reported by
the police, while in Eastern Europe police data are lower than those provided by health agencies to the
WHO.
24
These caveats notwithstanding, the WHO database provides sufficiently reliable data for our
present analysis.
First, it is interesting to look at how many gun-related deaths there are in Europe. Figures in the WHO
database indicate that between 2000 and 2012, more than 81,000 persons died from lethal firearms
wounds in the 33 European countries covered. Given the significant number of missing year-values, we
can estimate the total figure of firearms-related deaths in Europe in this period as approximately
94,000
I
. For the 27 EU Member States for which data is available,
II
this figure corresponds to almost
87,000 deaths. This means that within the EU, approximately 6,700 persons die each year as a result of
gunshot wounds. The European countries with the highest (age-standardized) death rates per 100,000
are Montenegro, Serbia, Finland, Cyprus, Croatia and France. The countries with the lowest death rates
per 100,000 are Romania, Poland, the United Kingdom, Spain and the Netherlands (see table 5).
I
In this calculation, the missing values for each country were replaced by the average number of firearms-related deaths in the
available years.
II
Data for Greece are not available.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 20
Table 5: Firearms-related deaths in 33 European countries, 2000-2012
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Average
/year
Age-
standardized
death rate per
100 000I
Austria
280
245
229
249
242
210
215
235
247
216
230
236,2
2.10
Belgium
330
309
315
256
210
231
189
197
254,6
1.60
Bulgaria
139
135
107
136
170
107
99
120
112,6
1.50
Croatia
200
160
153
140
165
162
139
146
148
145
133
103
120
147,2
2.37
Cyprus
9
11
12
11
13
15
7
7
24
12,1
2.59
Czech Republic
232
254
223
249
208
188
220
193
191
179
185
224
213
212,2
1.71
Denmark
102
89
96
70
99
92
79
65
67
75
86
71
53
80,3
0.84
Estonia
62
45
48
42
36
38
29
24
26
38
34
41
34
38,2
2.22
Finland
235
270
229
255
253
184
206
197
238
206
195
184
157
216,1
2.68
France
2,479
2,340
2,219
2,263
2,063
2,105
2,041
1,836
1,956
1,864
1,736
1,793
2,057.9
2.35
Germany
1,112
1,102
1,061
1,122
1,145
1,044
953
959
953
936
903
878
819
999.0
0.71
Hungary
117
136
124
112
112
105
85
87
95
85
95
91
95
103.0
0.83
Iceland
8
6
3
5
4
5
8
6
4
4
5.3
1.31
Ireland
41
52
57
46
49.0
1.00
Italy
966
807
799
793
769
798
780
816.0
1.07
Latvia
69
72
69
58
47
43
52
48
47
43
32
42
34
50.5
1.52
Lithuania
111
76
62
66
66
63
45
38
47
50
53
37
39
57.9
1.21
Luxembourg
11
9
13
6
6
9
11
14
11
7
11
6
9
9.5
1.54
FYR Macedonia
50
49
35
34
38
41.2
1.85
Malta
8
5
2
2
7
1
4
8
7
6
9
2
6
5.2
1.52
MontenegroII
68
49
57
54
57.0
8.20
Netherlands
115
114
105
104
100
104
82
82
78
93
76
96
92
95.5
0.55
Norway
126
109
106
115
103
87
98
67
88
107
87
142
88
101.8
1.62
Poland
211
160
182
154
141
121
97
116
114
116
98
99
104
131.8
0.25
Portugal
221
208
158
160
152
188
142
141
171.3
1.12
Moldova
65
76
57
53
49
33
30
34
21
21
26
37
29
40.8
0.73
Romania
76
59
67
62
51
57
48
35
34
37
42
32
34
48.8
0.15
Serbia
326
350
290
303
276
271
258
289
257
285
284
263
254
285.1
3.03
Slovakia
155
122
128
131
145
113
0
0
83
106
95
98.0
1.62
Slovenia
73
48
65
80
67
61
56
54
61
44
50
59.9
1.95
Spain
331
308
324
350
310
278
293
267
307
322
288
257
261
299.7
0.51
Sweden
169
164
192
141
156
137
139
135
138
157
138
135
141
149.4
1.48
United Kingdom
184
169
174
182
167
199
136
164
138
155
166.8
0.25
Source: European Detailed Mortality Database (WHO)
25
Secondly, we can observe a downward trend in the number of firearm related deaths in Europe.
Between 2000 and 2012 the annual number of firearms-related deaths decreased by approximately
20%. With the exception of Cyprus and Ireland, a decrease in the total number of firearms-related
deaths can be observed in all the European countries covered. In larger countries such as France and
I
For the most recent available year
II
In the WHO database, the number of firearms deaths in Montenegro for 2000 and 2006 are given as 0. Since these numbers
are most probably not correct, we exclude them from this table and further analyses.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 21
Germany, for example, the total number of firearms-related deaths decreased by as much as 26-28%.
While a general downward trend can be observed, this trend seems to have been especially strong
among Central and Eastern European countries. In countries such as Estonia, Latvia, Lithuania, Poland,
Moldova and Romania, the total number of firearm-related deaths halved from 2000-2012.
Third, our analyses of WHO data show that the vast majority of firearms-related deaths in Europe are
the result of successful suicide attempts. Approximately three-quarter of the reported gun deaths in this
period were suicides: of the 81,556 recorded cases of gun deaths in Europe in 2000-2012, almost
60,0000 were cases of suicide. Gun suicides were the most important form of gun deaths in almost all
European countries: only in the Netherlands, FYR Macedonia and Moldova could a higher number of gun
homicides than gun suicides be observed. With 12,016 recorded cases in the same period, homicides
accounted for 15% of the firearms-related deaths in Europe (see tables 13 and 15 in Annex). The other
firearms-related deaths were the result of accidents, or cases in which the cause of death could not
easily be determined. In the remaining paragraphs of this section, we will go more deeply into the
findings of our analyses of available data on gun suicides and gun homicides.
2.1 Suicide
Based on the available WHO data, we can reliably estimate the total annual number of gun suicides in
the 33 European countries concerned to be approximately 5,250, of which 5,000 take place within the
EU. Of all successful suicide attempts in Europe, approximately 9% are undertaken with a firearm.
Significant differences can be discerned between different European countries in this respect. In
countries such as Montenegro, Cyprus, Austria, Finland and Norway the relevant percentage is 16-38%,
while in countries such as Poland, Moldova, Romania and the Netherlands the percentage is 2% or lower
(see table 13 in Annex). In contrast to the general strong decline in total numbers of gun deaths, no
similar trend can be seen with regard to gun suicides. While the number of gun suicides decreased
spectacularly over the reporting period in France and Germany (see table 13 in Annex), this was not a
general trend in Europe. In most European countries a decrease could be observed, but this decrease
was often rather small. In addition, the available data indicates that in countries such as Bulgaria, Cypr us
and Portugal the number of gunshot suicides increased quite strongly in the later years of the period.
This finding of divergent national trends in the number of gun suicides is not surprising and confirms the
results of earlier studies.
26
According to the WHO data, the countries with the highest (age-standardized) gun suicide rates per
100,000 were Montenegro, Finland, Serbia, Austria and Croatia, while the countries with the lowest gun
suicide rates were Romania, Poland, the United Kingdom, Moldova and the Netherlands (see figure 6).
FIREARMS AND VIOLENT DEATHS IN EUROPE P 22
Figure 6: age-standardized gun suicide rates per 100,000 persons in 33 European countries, most
recent available year
Source: European Detailed Mortality Database (WHO)
27
0.06
0.10
0.16
0.22
0.24
0.25
0.33
0.56
0.58
0.60
0.67
0.67
0.77
0.78
0.87
0.87
0.96
1.03
1.06
1.11
1.15
1.31
1.31
1.37
1.49
1.78
1.80
1.88
1.93
1.96
2.05
2.31
6.33
0 1 2 3 4 5 6 7
Romania (2012)
Poland (2012)
United Kingdom (2010)
Moldova (2012)
Netherlands (2012)
Malta (2012)
Spain (2012)
Ireland (2010)
Germany (2012)
FYR Macedonia (2010)
Italy (2011)
Hungary (2012)
Denmark (2012)
Portugal (2012)
Slovakia (2010)
Bulgaria (2012)
Lithuania (2012)
Sweden (2012)
Latvia (2012)
Belgium (2010)
Luxembourg (2012)
Iceland (2009)
Cyprus (2012)
Czech Republic (2012)
Norway (2012)
Estonia (2012)
France (2011)
Slovenia (2010)
Croatia (2012)
Austria (2012)
Serbia (2012)
Finland (2012)
Montenegro (2009)
FIREARMS AND VIOLENT DEATHS IN EUROPE P 23
The international literature suggests that it is mainly men who use firearms to commit suicide, and this
is confirmed by the European WHO data. The overwhelming majority of gun suicides in Europe over the
period involved men (96%), while the number of women using a firearm was limited. Most gun suicides
were perpetrated by older men: more than two thirds of all male suicide victims were at least 50 years
old and almost half (42%) of all male suicide victims were aged at least 65 (see figure 7). Those few
women who used a firearm to commit suicide were generally somewhat younger: half of them were
between 30 and 55 years old. While in the United States much attention has been given to minors
committing suicides with firearms, youth suicides with firearms are not common in Europe. When
analysing the European WHO data we find that 71 gunshot suicides victims younger than 20 years old
were recorded in Europe in the most recent available year. This group therefore represented only 1.3%
of all gun suicide victims. More information on the context in which these gun suicides took place is
lacking in the WHO data. Previous research has consistently demonstrated that the shots are usually
aimed at the head (mouth, temple, forehead) or the chest.
28
In addition, research has indicated that gun
suicides most frequently take place at home,
29
that the suicides often had alcohol in their blood,
30
and
that a majority were neither in contact with psychiatric services nor had a history of previous self-
harm.
31
Figure 7: age-distribution of male and female gunshot suicide victims, Europe, most recent year
Source: European Detailed Mortality Database (WHO)
32
A comparison of the results of previous studies on gun suicides in specific European countries (Denmark,
Italy, England, Germany, Scotland and Sweden)
33
shows rather inconsistent findings with regard to the
type of gun used. Although in most countries shotguns were the main type of gun used to commit
suicide, the distribution of the types of guns used for the purpose differed from country to country. The
WHO database does not contain reliable information on the type of firearm used to commit suicides for
all European countries covered. Although distinct categories are provided for suicides using a (1)
handgun, (2) rifle, shotgun or larger firearm, and (3) other or unspecified firearms, for most
0%
2%
4%
6%
8%
10%
12%
14%
16%
<10 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+
Male victims Female victims
FIREARMS AND VIOLENT DEATHS IN EUROPE P 24
European countries the vast majority of suicides are categorized in the latter category. We analysed
the distribution of the types of firearms used to commit suicide for the five most recent available years
for the 33 European countries included in the WHO database (see table 14 in Annex) and found only 11
countries where the share of “other or unspecified firearm discharge” was below 50%.
An analysis of the types of guns used for suicide in these 11 countries shows a remarkable difference.
For the Nordic and Western European countries (Denmark, Finland, Iceland, Ireland, Sweden and the
United Kingdom), the data suggest that long guns such as shotguns or rifles were used in the
overwhelming majority of successful gunshot suicides, while for the other countries (Bulgaria, Czech
Republic, Lithuania, Serbia, Slovakia), handguns were most frequently used in successful gunshot
suicides (see table 6).
Table 6: Types of firearms used for gun suicides
I
in 11 selected European countries (over the five most
recent available years)
Handgun
discharge
Rifle/shotgun/
larger firearm
discharge
Other and
unspecified
firearms
discharge
Total
Bulgaria (2008-2012)
225
26
97
348
%
65%
7%
28%
100
Czech Republic (2008-2012)
326
74
358
758
%
43%
10%
47%
100
Denmark (2008-2012)
69
216
21
306
%
23%
71%
7%
100
Finland (2008-2012)
283
541
47
871
%
32%
62%
5%
100
Iceland (2005-2009)
0
17
7
24
%
0%
71%
29%
100
Ireland (2007-2010)*
5
51
47
103
%
5%
50%
46%
100
Lithunia (2008-2012)
43
23
60
126
%
34%
18%
48%
100
Serbia (2008-2012)
550
131
243
924
%
60%
14%
26%
100
Slovakia (2006-2010)
91
33
28
152
%
60%
22%
18%
100
Sweden (2008-2012)
88
372
140
600
%
15%
62%
23%
100
United Kingdom (2006-2010)
8
303
250
561
%
1%
54%
45%
100
* For Ireland only data from 2007-2010 is available
Source: European Detailed Mortality Database (WHO)
34
, own calculations
I
We used ICD10 codes x72 (Intentional self-harm by handgun discharge), x73 (Intentional self-harm by rifle, shotgun and larger
firearm discharge) and x74 (Intentional self-harm by other and unspecified firearm discharge)
FIREARMS AND VIOLENT DEATHS IN EUROPE P 25
Comparing these results with the analysis of types of firearms possessed in the different countries
according to the ICVS (see table 1) strongly suggests that the types of firearms used to commit suicides
across European countries reflect the general ownership rates of the different types of firearms in these
countries. In countries with rather lower percentages of handgun possession, we tend to observe low
percentages of the use of a handgun in gunshot suicides.
A relationship between gun possession and gun suicides can also be deduced from other findings of
the rather limited number of European studies focusing on gun suicides. The results of a UK study,
for example, indicate that a large majority of the firearms used were owned by the deceased. In
the other cases, the firearm was most often acquired within 24 hours of death and usually
belonged to relatives or a friend.
35
Despite the differences in types of guns used for suicide, the
studies that looked into the legal status of the firearms used generally indicate that most gunshot
suicides are committed with legally-held firearms.
36
In addition, some studies have found that a
significant share of guns used were service weapons
37
or were perpetrated by persons with
professional access to firearms
38
. Information on the legal status of the firearm used is not available
in the WHO data and can therefore not be analysed in this report.
2.2 Homicide
In a recent study the United Nations Office on Drugs and Crime (UNODC) estimated that in 2012
approximately 437,000 persons across the world died as the result of an intentional homicide in a non-
conflict situation. This comes down to 6.2 homicide victims per 100,000 persons. There are, however,
large differences in homicide rates between regions and sub-regions. The homicide rate is Europe is
considered among the lowest in the world
39
and this is probably one of the main reason for the relative
scarcity of homicide research in Europe. Research on homicides has long been dominated by scholars
from the United States. An increasing amount of European homicide research has, indeed, been
undertaken, often examining whether US-based findings hold good in Europe as well. But much of this
research has been relatively concentrated in specific countries with strong homicide research traditions
such as the United Kingdom, the Netherlands and Finland, or has focused on specific sub-types of
homicide such as intimate partner homicides or murder-suicides.
40
This means that rather little
information is available on gun homicides across Europe.
Analysis of the WHO data indicates that between 2000 and 2012, European countries registered more
than 12,000 homicides committed using a firearm. From these figures we can estimate the total number
of gun homicides in this period to be approximately 15,000 in Europe and 13,200 within the EU (see
table 15 in Annex). This comes down to 1,500 gun homicides per year in Europe and 1,000 gun
homicides per year within the EU. The WHO data indicate that a gun was used in approximately 20% of
all successful homicide attempts in Europe. This is a rather low proportion, give that firearms are the
FIREARMS AND VIOLENT DEATHS IN EUROPE P 26
most widely used weapons for homicides on a global level (41%).
41
The European countries with the
highest shares of gun homicide in the total numbers of homicides were Montenegro (93%), Cyprus
(63%), FYR Macedonia (50%), Italy (45%) and Ireland (42%).
The chance of becoming a victim of gun homicide differs strongly across Europe (see figure 8). The
European countries with the highest firearm homicide rates per 100,000 inhabitants are Montenegro,
Cyprus, and FYR Macedonia. The European countries with the lowest firearm homicide rates per
100,000 inhabitants are Iceland, Luxembourg, Malta, the United Kingdom, Poland, Slovenia, Austria,
Denmark and Germany.
Figure 8: age-standardized gun homicide rates per 100,000 persons in 33 European countries, most
recent available year
Source: European Detailed Mortality Database (WHO)
42
Interestingly, as with the total number of gun deaths, a general downward trend can also be observed in
the number of gun homicides in Europe: of the 30 countries with a total of more than 5 gun homicides
in 2000, 27 experienced a significant decrease in gun homicides by 2012. In countries such as Germany
and Spain, for example, the number of gun homicides decreased from respectively 101 and 105 in 2000
0.00
0.00
0.00
0.04
0.05
0.05
0.05
0.06
0.06
0.07
0.10
0.11
0.11
0.12
0.14
0.16
0.17
0.17
0.23
0.24
0.25
0.30
0.32
0.33
0.33
0.34
0.34
0.37
0.39
0.61
1.02
1.28
1.87
0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00
Iceland (2009)
Luxembourg (2012)
Malta (2012)
United Kingdom (2010)
Poland (2012)
Romania (2012)
Slovenia (2010)
Austria (2012)
Denmark (2012)
Germany (2012)
Lithuania (2012)
Hungary (2012)
Norway (2012)
Spain (2012)
Czech Republic (2012)
Estonia (2012)
Slovakia (2010)
Sweden (2012)
France (2011)
Latvia (2012)
Portugal (2012)
Netherlands (2012)
Finland (2012)
Ireland (2010)
Belgium (2010)
Italy (2011)
Bulgaria (2012)
Moldova (2012)
Croatia (2012)
Serbia (2011)
FYR Macedonia (2010)
Cyprus (2012)
Montenegro (2009)
FIREARMS AND VIOLENT DEATHS IN EUROPE P 27
to 61 in 2012. An especially spectacular decline can be observed in Central and Eastern European
countries and the Baltic States. In countries such as Croatia and Poland the number of gun homicides
decreased from respectively 48 and 79 in 2000 to 17 and 20 in 2012, while in Estonia, Latvia and
Lithuania we can observe a decrease from respectively 23, 31 and 51 gun homicides in 2000 to 2, 5 and
3 gun homicides in 2012 (see table 15 in Annex). Only in France, FYR Macedonia and Ireland does a
slightly different picture emerge: in the first two countries the decline was rather limited, while in the
latter a slight increase in the (rather small) number of gun homicides can be observed in recent years.
The decrease in gunshot homicides in Europe does not come as a surprise, since the given period also
showed a general decline in overall homicides rates across Europe.
European studies on gun homicide indicate that men are more often the victims of homicide than
women.
43
The high proportion of men among homicide victims is confirmed by the European WHO data:
in all European countries the number of male gun homicide victims was higher than the number of
female gun homicide victims. In total 76% of all homicide victims in Europe over the period were men.
Our analysis of the WHO data further indicates that the risk of becoming a homicide victim increases
significantly when men enter their early twenties, and this homicide risk reaches its peak for them
between the age of 35 and 39. Half of all male homicide victims over the period were between 25 and
44 years old. The female homicide risk, on the other hand, reaches its peak between the age of 40-49.
While younger men are particularly vulnerable for homicide, most female homicide victims are older
women (see figure 9). In the UK, a trend towards increasingly younger victims of firearm-related
violence has been observed;
44
but it is unclear to what extent this trend can be generalized to other
European countries.
Figure 9: age-distribution of male and female gunshot homicide victims, Europe, most recent year
Source: European Detailed Mortality Database (WHO)
45
Some research has been undertaken into the types of firearms used for gun homicides in Europe, but
they have yielded divergent results.
46
In the WHO database, reliable information on the type of firearm
used in gun homicides in Europe is lacking for most European countries. There are only seven European
0%
2%
4%
6%
8%
10%
12%
14%
16%
<10 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+
Male victims Female victims
FIREARMS AND VIOLENT DEATHS IN EUROPE P 28
countries for which the WHO data show a proportion of “other or unspecified firearm discharge” lower
than 50%. With the exception of Slovakia (which recorded only a small number of gunshot homicides),
we can observe that in all of these countries more homicides were committed with handguns than with
shotguns or rifles (see table 7). Even though most countries in this table are Central European ones, this
finding suggests that different types of firearms are generally used for homicides than for suicides.
Comparing the results of analyses on the types of firearm used for homicides and suicides for all
available European countries (see tables 14 and 16 in Annex), we can observe that even in countries
with high ratios of shotgun/rifle use for suicides, handguns are more regularly used for homicides.
Table 7: Types of firearms used for gun homicides
I
in seven selected European countries (for the five most
recent available years)
Handgun
discharge
Rifle/shotgun/lar
ger firearm
discharge
Other and
unspecified
firearms
discharge
Total
Bulgaria (2008-2012)
86
17
45
148
%
58%
11%
30%
100
Czech Republic (2008-2012)
49
6
25
80
%
61%
8%
31%
16
Denmark (2008-2012)
22
11
8
41
%
54%
27%
20%
6
Finland (2008-2012)
34
31
35
100
%
34%
31%
35%
100
Lithuania (2008-2012)
19
7
15
41
%
46%
17%
37%
100
Serbia (2008-2012)
143
22
72
237
%
60%
9%
30%
100
Slovakia (2006-2010)
9
12
8
29
%
31%
41%
28%
100
Source: European Detailed Mortality Database (WHO)
47
, own calculations
The UN Office on Drugs and Crime has identified three types of homicides: homicide related to other
criminal activities, interpersonal homicide, and socio-political homicide. These types of homicide take
place in very different contexts. Within the group of homicides related to other criminal activities, two
sub-groups can be distinguished. First are the homicides committed by organized criminal groups, which
are generally deliberate and aimed at achieving long-term goals, for example the elimination of rivals,
making a show of strength and/or securing territory. The perpetrators and victims of homicides related
to criminal activities are generally male. The second type are homicides committed while perpetrating
other conventional criminal activities such as a robbery. Although they are sometimes perpetrated in
order to accomplish the original crime and/or avoid detection, these homicides are generally less
deliberate and can be considered as collateral damage from the perpetrators’ viewpoint.
I
We used ICD10 codes x93 (Assault by handgun discharge), x94 (Assault by rifle, shotgun and larger firearm discharge) and x95
(Assault by other and unspecified firearm discharge)
FIREARMS AND VIOLENT DEATHS IN EUROPE P 29
Interpersonal homicides are generally linked to an interpersonal conflict; often involve intimate
partners or family members; and can be the result of a premeditated design or a more impulsive act of
violence. In contrast to the other types of homicide, the incidence of intimate partner/family violence is
rather stable on a global level. In addition, the female victim rates of these homicides are generally
higher than the male victim rates.
Socio-political homicides such as hate crimes or terrorist attacks are generally committed in the public
sphere and seen by the perpetrator as instrumental in pushing a social or political agenda. These
homicides are generally the result of premeditation and organization. In contrast to interpersonal
homicides, the perpetrator often does not know the victim; victims are chosen for what they represent
and/or for the message the homicide sends to the rest of the population.
48
The WHO data do not provide an insight into the distribution of firearms-related homicides between
these three different types, but an analysis of homicides in three European countries (Finland, the
Netherlands and Sweden) suggests that the majority of homicides in Europe are the fatal result of
interpersonal violence in the domestic sphere (perpetrated by the intimate (ex-)partner, or other family-
related violence).
49
The number of homicides in the criminal milieu varies from country to country, but
they occur much less frequently (see figure 10). Despite the recent terrorist attacks in Paris (January
2015) and Utøya (in July 2011), killing respectively 12 and 77 persons, socio-political homicides in
Europe do not occur very frequently and are thus not representative for most homicides in Europe.
50
Interestingly, gender differences have been frequently observed in the relationship between victim and
perpetrator: female victims are much more likely to be killed by an (ex-)partner of family member than
are men.
51
The pattern that often emerges is that women are typically killed following domestic
disputes, while men are killed in a wider array of contexts.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 30
Figure 10: Share of homicides, by typology, in Finland, the Netherlands and Sweden, 2003-2006
Source: European Homicide Monitor
52
The observation that younger men and older women in particular are vulnerable to gunshot homicides
(see figure 9) may be a reflection of the existence of different types of homicide with specific
demographics groups as their prime targets. One could argue, for example, that young men are at a
higher risk of becoming the victim of criminal-related homicides, while older women are particularly at
risk for homicides following domestic disputes; but until more data becomes available this remains a
hypothesis.
In addition, previous research suggests that different types of guns are used for different types of
homicides. A study on gun homicides in Oslo (Norway) and Copenhagen (Denmark) found that female
victims were typically shot with a shotgun in their own house by their partner, while male victims were
shot on different locations, most often with a handgun, for a variety of reasons such as revenge or
financial disputes and for the most part by an acquaintance.
53
Handguns seem to be more popular as a
weapon in criminal homicides. An analysis of all homicides occurring between 1988 and 2003 in the
southern Italian province of Apulia, where organized crime groups hold control over illegal business,
indicate that pistols were used in almost two thirds of cases where the type of firearm was known.
54
The
overwhelming majority of homicides in that region are motivated by battles for turf control between
rival organized crime groups, including juvenile gangs; in the context of a robbery; or in an attempt to
avoid arrest.
55
Research on the legal status of the firearms used in gun homicides in Europe is largely lacking. In
contrast to the findings of studies on gun suicides, the few European studies on gun homicides that have
analysed the legal status of the guns used found that the majority of them were illegally held.
56
Not all
36%
3%
3%
5%
53%
Finland
39%
19%
9%
5%
28%
The Netherlands
45%
12%
7%
10%
26%
Sweden
Domestic violence Criminal milieu Robbery Nightlife violence Other
FIREARMS AND VIOLENT DEATHS IN EUROPE P 31
European studies came to this conclusion, however. An older study on gun deaths in Denmark found
that the majority of gun homicides in Denmark were committed with legally held firearms (either freely
accessible weapons, or licensed firearms).
57
FIREARMS AND VIOLENT DEATHS IN EUROPE P 32
3 Gun Ownership and Violent Death: lessons from
previous research
In the previous paragraphs we looked at the available data on firearms ownership and the levels of
violent death in Europe. Based on these data, we will now review the results of previous research on the
possible associations between these two variables. The central question is whether the prevalence or
availability of guns in societies is positively or negatively linked to the rates of violent death (gun-related
and overall), or whether they are not related in any systematic way. In other words, we ask whether
more guns in society are linked to higher levels of violent death, or (to test a hypothesis sometimes put
forward by pro-gun lobbies) whether high levels of gun possession are related to low levels of violent
death.
Studies analyzing the relationship between firearms availability and violent death specifically for the
European context are relatively scarce. There is a body of literature on the links between gun possession
and violence but, as we have noted, most of it is American. Given the particularities of the gun situation
in the US, we should be careful in translating the results of this literature to the European context.
Nonetheless, this literature offers an interesting starting-point for seeking insight into the relationship
between gun availability and violent deaths. The existing studies originate from different disciplines and
use various research designs and methods. An important distinction can be made, for instance, between
individual-level and population-level studies studies. In the following paragraphs we will give an
overview of the main findings of both types of studies. In public debates, it often happens that only a
limited number of dispersed case studies are cited: which of course brings the risk of bias. To avoid this
risk, we will primarily focus on the findings of systematic reviews of available studies.
3.1 Individual-level versus population-level studies
Individual-level studies, often conducted by public health experts, relate gun possession by a person or
in a household to the risk of that person’s dying as a result of suicide or homicide. These studies are
generally case-controls studies in which a group of suicide or homicide victims is compared to a matched
group of persons who are either still alive or dead from other causes. The goal is to identify risk factors
for victimization. These controls are typically matched with carefully selected individual variables that
might be relevant, such as gender and different socio-economic variables, and result in a number of
probability ratios that reflect the increased risk of suicide or homicide when possessing a firearm at
home. Some of the limitations of this method arise from the non-random selection of cases or controls,
FIREARMS AND VIOLENT DEATHS IN EUROPE P 33
and from mis-classification of the outcome (suicide or homicide) or of the facts of gun ownership
I
. In
addition, the results may always be confounded by other variables that were not included in the
analysis.
58
In population-level studies the unit of analysis is not the individual but a larger community. These
studies typically assess the correlations between the rate of gun ownership and suicide and/or homicide
rates at the international or national level. As in individual-level studies, a number of unobserved and
complicating factors might be associated both with gun ownership and with the risk of falling victim to
violent death. Because these variables are not counted in, there is always the risk that research results
point to associations that may be spurious. Most population-level studies therefore try to limit this
danger by including several control variables such as rates of criminality, urbanization, poverty,
unemployment or depression.
59
In a recent study, Bangalore and Messerli looked at the relationship between guns and deaths in 27
developed countries.
60
They found a significant positive correlation between the number of guns per
capita per country and the rate of firearm-related deaths. These findings, however, have a limited range.
First, the authors did not check for correlations with overall violent death rates. Further, looking at
absolute levels of violent death is crucial. The overriding motivation of policy efforts to restrict the
availability of guns is, after all, to reduce deaths and injuries from all sources, not only through the use
of firearms. Moreover, Bangalore and Messerli’s findings have limited significance because the authors
did not distinguish between suicides and homicides, a distinction that is commonly considered
important in research on connections between guns and violent deaths. Over the last decades there
have been a number of studies that not only took into account the absolute levels of violent deaths, but
also looked at homicide and suicide separately. The following paragraphs offer an overview of the most
important findings of these studies, and elaborate on the dynamics that seem to be at play to explain
the relationship between gun availability, suicide rates and homicide rates.
3.2 Suicide
In a 2014 international review of individual-level studies, Anglemeyer and colleagues found that access
to firearms is associated with a higher risk for completed suicide. Among persons with access to
firearms, they found significantly increased odds of completed suicide compared with those without
access. Citing previous studies, they noted that the increased risk for suicide associated with firearms in
the house is not unique to persons with a history of mental illness, and may be more of an indicator of
the ease of impulsive suicide. In other words, guns in the home appear to convey a particularly
I
Stroebe illustrates this last point by pointing to the result of a study on the validity of self-reported data on the presence of a
firearm in the household, which suggested that 11% of the responses by household members on gun possession were invalid.
Stroebe, W (2013), ‘Firearm possession and violent death: A critical review’, Aggression and Violent Behavior, 18 (6), p.709-
721.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 34
increased risk for suicide in persons with no apparent psychopathology. They add that the way weapons
are kept plays a role: firearms that are stored loaded or unlocked are more likely to be used than those
that are unloaded or locked away. Anglemeyer and colleagues concluded that the evidence they
reviewed highlighted the risks of having a firearm in the home, and that restricting that access might
effectively prevent injury.
61
A number of population-level studies have also focused on the potential relationship between gun
possession and suicide rates. A multi-level modelling of the available international longitudinal data by
Ajdacic-Gross and colleagues confirms the effect of gun availability on firearms suicides: the proportion
of households possessing firearms overwhelmingly influences gun suicide rates.
62
Looking at the societal
level, David Hemenway points out that evidence from many population-level studies of guns and violent
deaths is as strong as that from individual-level studies. Mainly reviewing US studies, he notes that
across US cities, states, and regions, higher levels of household gun ownership are associated with
higher firearm-related and overall suicide rates. He adds that these studies show that adults in
households with firearms are no more depressed or suicidal than those in households without firearms,
yet they are far more likely to die of suicide.
63
In Europe a limited number of population-level studies
have been conducted to investigate the relationship between gun availability and firearm suicides. In
2010, Ajdacic-Gross and colleagues found that Swiss cantons where firearms ownership in private
households is more frequent also tend to have higher proportions of firearms suicides.
64
In Austria,
Etzersdorfer et al. found a strong correlation between the average number of gun licences issued in the
nine Austrian Länder (1990-2000) and gunshot suicides. The number of licences was only weakly
correlated, and for some of the years under investigation negatively correlated, with other methods of
committing suicide and with the suicide rate in general.
65
The link between the availability of a weapon and committing suicide, among other things, can also be
deduced from findings in studies on gunshot suicides with regard to the type of firearm used. First of all,
a number of studies have demonstrated that the proportion of handguns used for gunshot suicides is
much lower in places where they are less available because the legal acquisition of these firearms is
limited (Avis, 1994). Secondly, some citizens have easier access to (certain types of) firearms because of
their professional activities. This difference in access is reflected in the type of weapon used to commit
suicide. A study on gun-shot suicides in Northern Ireland, for example, found an interesting difference
with regard to the types of firearms used by security forces and by civilians who committed suicide.
While civilians mostly used shotguns to commit suicide (78%), members of the security forces mostly
used handguns (84%).
When comparing the results of cross-national studies, the National Research Council, a research body
linked to the US National Academies, concluded that a consistent association between gun ownership
and the share of gun suicides exists, but in contrast to US studies, the cross-national studies do not
reveal a consistent association between gun ownership and overall suicide rates.
66
The results of two
FIREARMS AND VIOLENT DEATHS IN EUROPE P 35
often-cited studies illustrate this pattern of non-consistent findings. Using data from the 1989
International Crime Survey on gun ownership and WHO data on suicides for 14 countries, Killias found
significant positive correlations between the rates of household gun ownership and the national rates of
suicide (with a gun and overall), as well as with the proportion of suicides committed with a gun. This
would mean not only that the more guns there were in a country, the higher the suicide rates were, but
also that more of these suicides were committed with a gun. The study showed no negative correlation
between the rates of gun ownership and the rates of suicide committed by other means. This, according
to Killias, indicated that the other means were not used to substitute for the absence of guns in
countries with a lower rate of gun ownership.
67
When this study was updated in 2001 by Killias, Van
Kesteren and Rindlisbacher, using data from the International Victimization Surveys of 1989, 1992 and
1996, the results were different: while very strong correlations between the presence of guns in the
home and suicide committed with a gun were found, only non-significant correlations were found for
overall suicide rates.
68
The results of the above-mentioned studies consistently demonstrate a clear link between gun
possession rates and gun-related suicides, and some studies suggest a link between gun possession
rates and the overall suicide rate. Going beyond merely analyzing connections between gun availability
and violent death, a number of studies have looked more in detail at the dynamics that seem to play a
role in explaining these relationships. With regard to suicide, an important distinction should be made
between suicide intent and suicide risk. Several factors play a role in the outcome of a suicide attempt,
including the degree of intent and the lethality of the method used. In this section we will elaborate on
the different dynamics that seem to be at play. Two key concepts deserve particular attention: the
lethality of firearms, and the degree of impulsiveness on the perpetrator’s part.
A first factor we need to take into account is that using a firearm is generally considered to be the most
lethal method for committing suicide. Traditionally, much of the literature on suicide risks has
emphasized the lethality of the chosen method. Individuals generally have a preference for a specific
suicide method, formed by an interplay of personal characteristics and cultural factors. A number of
more context-specific aspects will also play a role, such as the availability of certain means, the technical
skills needed, the degree of disfigurement after death, or the symbolism of choosing a particular way to
die.
69
An analysis of the success rate of different suicide methods in the United States demonstrates that
the use of a firearm is by far the most lethal for committing suicide: while 85% of all suicide attempts
with a firearm are fatal, this percentage drops to 69% for hanging and 31% for jumping off a high
building (see table 8). The highly lethal nature of firearms-related suicide attempts has been confirmed
by other studies, and it was has been shown that the lethal nature of gunshot suicide attempts does not
differ significantly as between men and women or between adults or minors.
70
FIREARMS AND VIOLENT DEATHS IN EUROPE P 36
Table 8: Overview of percentage of fatal injuries resulting from an attempted suicide by different
methods, United States, 2001 (N=353,613)
Method
% fatal
Firearm
85%
Suffocation/inhalation
69%
Fall
31%
Poisoning/overdose
2%
Cutting/piercing
1%
Other
3%
Unspecified
7%
Total
9%
Source: Vyrostek, Annest & Ryan
71
After reviewing almost 100 studies evaluating different suicide prevention strategies, Mann and
colleagues confirm that suicide attempts using highly lethal methods result in higher death rates. They
further conclude that the suicide rates under study decreased after restricting access to these means,
for example by placing restrictions on pesticides, by constructing barriers at jumping sites, or by
increasing firearm controls. Where the use of these lethal methods is common, such restrictions on
available means have led to lower overall suicide rates.
72
A second factor we need to take into account is the nature of a suicide attempt, in particular the
question of whether it results from an impulsive decision or reflects a longer history of psychiatric
treatment or previous suicide attempts. In their study on suicide prevention methods, Florentine and
Crane argue that “the fundamental assumption underlying attempts to limit access to suicide methods
is that, in many cases, periods of high suicide risk are relatively short and limiting access may delay an
attempt until the period of high-risk passes”.
73
It is clear that cases of suicide can vary strongly in the
degree of premeditation. Earlier research suggests that a significant proportion of suicide attempts are
impulsive acts.
74
According to American research, for example, more than two thirds of persons who
made near lethal suicide attempts took less than one hour between the decision to kill themselves and
the actual attempt. Almost a quarter took less than five hours.
75
Similar findings from other studies also
point to the short-time frame between the first thought of suicide and the actual attempt, for a large
number of suicide attempts.
76
A number of studies have indicated that the availability of firearms to impulsive persons may account
for high rates of gunshot suicides.
77
In their review of the UK literature on gunshot suicides, Haw et al
(2004) conclude that those who commit suicide with a firearm are less likely to have current or past
mental health problems, are less likely to have a history of self-harm, and are more likely to have had a
recent relationship dispute than people who committed suicide by other means.
78
Similar results were
found in a study on differences in characteristics of suicides using different methods in Finland: persons
who committed suicide by using a firearm less often had a history of psychiatric treatment or previous
suicide attempts. Their suicides were often associated with recent life events and with alcohol in the
FIREARMS AND VIOLENT DEATHS IN EUROPE P 37
blood at the time of death.
79
This finding is not surprising, since previous US studies demonstrated that a
significant share of the perpetrators of gunshot suicides had used alcohol or drugs and/or had
experienced an interpersonal conflict within 24 hours before the attempt, while only a small minority of
the perpetrators had a past history of suicide attempts.
80
Jackson and Marsh have noted that such
findings “supports the hypothesis that access to firearms is particularly relevant in cases in which the
suicide was impulsive and major risk factors were absent”. They conclude that “if they are determined,
people will be able to find the means to kill themselves”, but that “in practice it appears that as many
suicides are impulsive the availability of firearms does affect overall levels of suicide”.
81
US studies
suggest that impulsive behaviour is especially problematic with regard to potential youth suicides. More
particularly, an alcohol/gun-suicide connection seems to exist among adolescents, given that
adolescents who used a firearm to commit suicide were substantially more likely to have been drinking
than those who used a different method.
82
The lethality of the instrument of choice and impulsiveness seem to be mutually reinforcing dynamics in
explaining the outcome of suicide attempts: “Firearms are lethal instruments requiring relatively little
expert knowledge to operate them, so they may be the instrument of choice to those acting on impulse
who are seeking a certain and rapid death”.
83
The question that must raised here is whether a
substitution of means will occur when the availability of instruments of first choice is limited - for
example through policy intervention - and what the impact of this substitution will be on the lethality of
suicide attempts. After reviewing more than 40 different studies on the risk of method substitution with
regard to suicide, Daigle concluded that the risk of a preferred instrument being substituted by other
means seems small. In particular, individual-level studies demonstrate that many suicidal persons have a
preference for a specific method, and that limited access to this means by an individual will not
automatically lead to a shift towards another suicide method.
84
Other studies have acknowledged the
impact of limiting access to certain means on the short-term, but have stressed that this does little to
prevent suicidal behaviour in the longer term except by buying time for medical or psychosocial
interventions.
85
Although we know that the restriction of means does not address the root problem of many persons
who contemplate committing suicide, this restriction of means can play an important role in the
prevention of suicides. As Florentine and Crane correctly point out: “The fundamental assumption
underlying attempts to limit access to suicide methods is that, in many cases, periods of high suicide risk
are relatively short and limiting access may delay an attempt until the period of high-risk passes”.
86
Despite the possibility of substitution, limiting access to highly lethal means such as firearms can play an
important role in decreasing overall suicide rates: “It is recognized that suicidal crisis are usually of short
duration and that, if their fatal outcome is prevented, they will not be repeated or help will be made
available in the meantime. At worst, suicidal individuals deprived of their preferred means of committing
suicide will fall back on another, possibly less lethal, method”.
87
Reducing access to a highly lethal
FIREARMS AND VIOLENT DEATHS IN EUROPE P 38
method, such as a firearm, can thus decrease the total number of suicides even where substitution
occurs, since the proportion of people who survive the suicide attempts will increase.
Another factor we must take into account when trying to interpret the correlations found between gun
possession and gun suicides is that some people acquire a firearm for the specific purpose of committing
suicide or for killing someone else (reversed causality). A number of American studies have
demonstrated that the risk of suicide is significantly higher after a recent purchase of a firearm. Based
on a case-control study among members of a health maintenance organization in Washington State
(USA), Cummings and colleagues found, for example, that within the first year of the purchase of a
handgun by a family member the relative risk of committing suicide was more than five times higher
than that of the general population.
88
A similar observation was made by Wintemute and colleagues. In
addition, they found that the relative risk of suicide was the highest in the first week after the purchase,
and declined thereafter. This strongly suggests that a significant number of persons buy a firearm with
the specific intention to commit suicide with it.
89
Yet this reversed causality alone cannot entirely
explain the oft-demonstrated correlations, since several studies have found that minors - who are
generally not allowed to purchase firearms - are also at a higher risk of suicide if a firearm is present at
home.
90
Interestingly, the above-mentioned studies found that after the first year, the increased relative
risk for suicide persisted at a lower level for years. This corresponds with the hypothesis that a gun at
home may facilitate suicide in times of despair.
3.3 Homicide
Most studies on the possible linkages between gun possession and firearms related deaths have focused
on suicide. Sound research, and especially European studies, on the specific linkage between gun
ownership and homicides is more difficult to find.
The lack of good data and sound research is an important stumbling block in advancing our knowledge
on this issue. According to the UNODC, hypotheses about these linkages tend to fall into two categories:
“One suggests that easy access to firearms may facilitate the commission of homicide in a variety of
ways, including by fostering violent confrontations and by increasing their lethality, as well as, on a
different note, by facilitating the commission of crimes and the execution of targeted killings. The second
hypothesis suggests, on the other hand, that widespread availability of firearms may be a deterrent to
assault and aggressions, in that it may reduce the leverage and motivation of an armed perpetrator. A
number of methodological challenges, starting with the shortage of data on firearm availability, make it
difficult to provide definitive answers in either direction”.
91
Nonetheless, the results of a recent survey
among 85 authors of studies on firearms that were published in peer-reviewed journals indicates that
only 12% disagreed with the statement that “strong gun laws help reduce homicide”, while 71% of them
agreed. Among those who believed they were knowledgeable or very knowledgeable about this issue
FIREARMS AND VIOLENT DEATHS IN EUROPE P 39
the latter percentage increased to 77%. The support for this statement was most pronounced among
public health and medicine experts.
92
In their 2014 review of individual-level studies, Anglemeyer and colleagues found that access to firearms
is associated with a higher risk of being the victim of homicide. They found evidence for increased odds
of becoming a victim of homicide, certainly among women, when persons with and without access to
firearms were compared. Yet the association between firearm availability and homicide was found to be
more modest than that between firearm availability and suicide. With regard to homicide, the results of
previous population-level studies are not very consistent. Again we can illustrate this by mentioning the
two often-cited studies by Killias and colleagues, using the International Crime Survey data. In their first
study, using data from the 1989 International Crime Survey on gun ownership and WHO data on
homicides for 14 countries, significant positive correlations were found between the rates of household
gun ownership on the one hand and both national gunshot homicide rates and overall homicide rates on
the other. Similarly with the results on suicide, the proportion of homicides committed with a gun was
correlated with gun ownership rates, while no negative correlation was found between the gun
ownership and the rates of homicide committed by other means. This suggests the absence of a
significant substitution effect in countries with a lower rate of gun ownership.
93
In their updated study,
using data from the International Victimization Surveys of 1989, 1992 and 1996, the only correlation
found was between household gun ownership rates and gunshot homicide rates involving female
victims. For gun-related male homicide and overall homicide rates, no significant correlation was
found.
94
In the previous section we argued that in many cases the periods of high suicide risk are relatively short,
and that limiting access to lethal means such as firearms can therefore delay suicide attempts or drive
people towards less lethal means, resulting in lower overall suicide rates. A similar argument can be
made for limiting access to firearms as a way of preventing successful homicide attempts, since a
considerable proportion of homicides do not reflect a long-planned and deliberate rational choice to kill
someone, but rather, impulsive acts that can occur in high-stress situations. In homicide research a
distinction is often made between instrumental and expressive aggression. This distinction is regularly
considered fundamental for understanding aggressive behaviour and is often used in homicide research.
In instrumental aggression, the aggression is not an end in itself but is motivated by gain, and occurs
when a person feels that an act of violence is necessary to achieve another specific goal - for instance a
robbery that results in homicide. This aggression has been characterized as intentional, rational and
often premeditated. Expressive aggression, on the other hand, is often an emotional response to
frustration and actually motivated by the urge to harm someone. This type of aggression is generally
characterized by impulsiveness, emotions, passion, anger and lack of control. This instrumental-
expressive continuum has been further expanded upon by several authors.
95
FIREARMS AND VIOLENT DEATHS IN EUROPE P 40
There are strong reasons to suspect that a significant share of all gun homicides in Europe are the result
of expressive violence, especially when domestic disputes are involved. Homicide figures from the UK
for 2009/10, for example, indicate that approximately half of all homicides resulted from a quarrel, a
revenge attack or a loss of temper. When the perpetrator and victim were acquaintances, this
percentage rose to 60% compared to 38% when they were strangers to one another.
96
Previous studies
have found that domestic violence is often an act of expressive violence, since it is frequently
characterized by a loss of temper, impulsiveness and/or emotional reactions to (real or perceived)
provocation. Homicides that take place outside of the domestic sphere are more often characterized by
premeditation, calculation, control and the absence of emotional arousal.
97
In particular, homicides in
the criminal milieu, or those related to carrying out other felonies such as robbery or sexual assault, are
believed to be related to these latter characteristics and are thus often considered as primarily
instrumental acts of violence.
We believe that the different types of homicide are linked not only to different types of violence, but
also to different types of weapons. American research, for example, has indicated that the use of
firearms is more likely in homicides committed by acquaintances and strangers compared with
homicides committed by partners or family members.
98
This probably partially explains the consistent
research finding that gun ownership rates, which generally tend to reflect the number of legally held
firearms, are especially correlated with the risk for women of becoming the victim of homicide. While
female victims of homicide can usually be placed in a context of domestic disputes or other types of
interpersonal violence, male victims are generally more likely to be found in a criminal context. Further,
the results of a recent Finnish study on the choice of weapons used in lethal or potentially lethal
violence against parents indicated that adolescent perpetrators were more likely than adult
perpetrators to use a firearm in lethal or potentially lethal violence against their fathers. This was
explained by the physical strength hypothesis which suggests that, given that adults are generally
stronger than adolescents, using a firearm is a more rational choice than choosing a method that
requires physical contact.
99
FIREARMS AND VIOLENT DEATHS IN EUROPE P 41
4 Gun Ownership and Violent Deaths in
Europe: A New Analysis
Studies investigating the correlations between gun ownership and violent deaths in a purely European
context seem to be lacking. For the present report, we collected information on gun ownership and
violent deaths for 33 European countries
I
in order to calculate and analyse the correlations between
these variables in a European perspective. For the rate of gun ownership we used the estimates of
civilian firearms ownership (guns per 100 people) made by the Small Arms Survey in 2007
100
.
II
The data
on violent deaths were taken from the European Detailed Mortality Database of the WHO Regional
Office for Europe.
101
Given that the estimates of gun ownership were made in 2007, we chose to use the
death rates from 2007 as well. Additional variables for possible complicating factors such as
unemployment rates, alcohol consumption or criminality rates were not available, and could therefore
not be included in our analysis. This means we were not able to control for possible confounders. The
results of our analysis are given in table 9.
A first observation is that our analysis indicates a strong positive correlation between the gun ownership
rate in a country and the rate of firearms-related deaths (model 1, R=.527, p=.002). This means that in
Europe, the countries with the highest rates of gun ownership generally also have the highest rates of
gun deaths. The association with gun ownership is particularly strong when looking at male gun deaths
(R=.562, p=.001). For female gun deaths we also observe a significant correlation, but weaker than for
male gun deaths and only statistically significant when the outlier (in this case Montenegro) was
excluded (R=.412, p=.019). This gender difference can probably be attributed to the fact that most gun
deaths in Europe are the result of suicide and that gunshot suicides are primarily committed by (older)
men.
When looking at the specific results for suicide (model 2), we can observe a very strong positive
correlation between gun possession rates and gunshot suicides by male victims (R=.538, p=.001). Not
surprisingly given the high share of male suicides in the overall suicide rates (96%), the results of our
analyses indicate a strong positive correlation between gun possession rates and the total number of
gun suicides in Europe (R=.495, p=.003). The correlation between gun possession rates and female gun
suicides is weaker, and only statistically significant when the outlier (again Montenegro) was excluded
I
Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Iceland,
Ireland, Italy, Latvia, Lithuania, Luxembourg, FRY Macedonia, Malta, Moldova, Montenegro, Netherlands, Norway, Poland,
Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, United Kingdom.
II
For this analysis we preferred to use the Small Arms Survey estimates for gun ownership over the ICVS data because data for
more countries was available and more recent in the Small Arms Survey estimates.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 42
(R=.371, p=.037). This observation might be attributed to the small number of women committing
suicide with a firearm in Europe.
The correlation established in model 2 between national gun ownership rates and national gun suicide
rates in Europe is not surprising, and in line with the results of previous research using different data
sources (see chapter 3). In countries where people can easily access firearms, there will be more
suicides in which firearms are used. An important question that now arises is whether gun possession
rates are also connected to overall suicide rates. Our analyses show there is no statistically significant
correlation between gun possession rates and overall suicide rates (R=.008, p=.96), neither for male
suicides (R=-0.075, p=676) nor for female suicides (R=0.237, p=.185). At first sight this could be
interpreted as a sign that substitution of means is taking place: that is, that persons in countries where
less firearms are less easily available will use other means to commit suicide. A closer look at model 2,
however, indicates that neither for men (R=-.201, p=.575) nor for women (R=.221, p=.216) can a
significant correlation be found between gun possession rates and the rates of suicide in which no
firearm was used. This is an important finding with regard to substitution of means. If there was in fact a
considerable degree of substitution, one would expect that the rate of suicides committed with other
means than firearms would be significantly higher in countries with low gun ownership rates than in
countries with high gun ownership rates: but this is not the case. This suggests that means substitution
sometimes does occur in suicide attempts, but not very frequently and not with a similarly lethal
outcome. The more plausible explanation for the lack of significant correlations between national gun
possession rates and overall suicides in Europe is the fact that gun suicides only account for a small
proportion (9%) of all suicides in Europe (see chapter 2.1). While a strong impact of ownership on gun
suicides is obvious (and there is no indication for a substantial substitution effect), the share of gun
suicides in overall suicide rates is just too small to have a sizeable (and statistically measurable) effect on
overall suicide rates across Europe.
In model 3 we analyse the correlations between gun ownership rates and homicide rates. In this model
hardly any significant correlations were found. Only for the levels of female victims in gunshot
homicides was a (barely statistically significant) correlation with gun possession rates found, and this
only when the outliers were excluded from the analysis (R=.361, p=.042)
I
. This observed significant
correlation between gun possession rates and homicide rates involving female victims is in line with
previous research (see chapter 3.3). Since female homicide victims are generally the result of fatal
domestic disputes (see chapter 2.2), we can therefore assume that the impact of gun ownership on gun
homicides is greatest when dealing with the more expressive forms of aggression that are typically
connected with domestic disputes.
I
When the female gunshot homicide rates were correlated with the ICVS data on gun ownership a statistically significant
correlation was found.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 43
Table 9: correlations between gun ownership (Small Arms Survey) and violent deaths in 33
European countries
I
Pearson’s R
with outliers
Pearson’s R
without outliersII
Model 1: Firearms-related mortality (2007)
- Total
.527
P=.002
.681°
P=.000
- Total – men
.562
P=.001
.684°
P=.000
- Total – women
.268
P=.132
.412°
P=.019
Model 2: Suicide (2007)
- Total
.008
P=.966
Idem
- Total – men
-.075
P=.676
Idem
- Total – women
0.237
P=.185
Idem
- With gun
.495
P=.003
.678°
P=.000
- With gun – men
.538
P=.001
.673°
P=.000
- With gun – women
.166
P=.355
.371°
P=.037
- Without gun
-.101
P=.575
Idem
- Without gun – men
-.201
P=.261
-.070°°
P=.702
- Without gun –women
.221
P=.216
Idem
Model 3: Homicide (2007)
- Total
-.187
P=.298
.163°°°
P=.397
- Total – men
-.200
P=.264
.103°°°
P=.595
- Total – women
-.134
P=.457
.229°°°
P=.231
- With gun
.297
P=.093
.302°°°°
P=0.093
- With gun – men
.255
P=.152
Idem
- With gun – women
.344
P=.050
.361°°°°
P=.042
- Without gun
-.235
P=.188
-0.019°°°
P=.921
- Without gun – men
-.241
P=.178
-.080°°°
P=.681
- Without gun –women
-198
P=.269
.088°°°
P=.649
° Excluding Montenegro °° Excluding Lithuania °°°Excluding Estonia, Latvia, Lithuania & Moldova °°°° Excluding FRY Macedonia
I
Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Iceland,
Ireland, Italy, Latvia, Lithuania, Luxembourg, FYR Macedonia, Malta, Moldova, Montenegro, Netherlands, Norway, Poland,
Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, United Kingdom
II
To calculate outliers the following formula was used: Y<(Q1 – (2.2 x (Q3-Q1))) or Y>(Q3+ (2.2 x (Q3-Q1)))
FIREARMS AND VIOLENT DEATHS IN EUROPE P 44
The results of this analysis should be considered in the light of a number of limitations. First, for the
variable of gun ownership we used the results of the expert estimates made by Small Arms Survey. As
mentioned in chapter 1, these figures need be treated with care, and analyses using these figures must
accordingly be interpreted with caution. Secondly, our data on firearms do not specify the types of
firearm used in lethal incidents, their legal status, nor the way these weapons were stored. Integrating
more detailed information on type, legal status and storage modalities of firearms should enhance our
understanding of the relationship between gun ownership and violent deaths. Thirdly, although results
from population-level correlational studies are frequently used in arguments about the causal impact of
firearms availability on violent deaths, we must remain cautious about answering questions of causality
merely on the basis of this type of research. The level of lethal violence in society is affected by many
factors: besides the potential effects of gun availability, violent death rates are impacted by numerous
socio-economic and psychological factors such as income inequality, unemployment rates, poverty,
mental illness, divorce rates and so forth. As these factors vary across countries, they are important for
explaining why overall suicide and homicide rates vary widely across different European countries (see
figures 6 and 8); but we did not control for such variables in our analysis.
These limitations notwithstanding, the results of our analyses indicate the existence of significant
correlations between gun ownership and 1) gun suicides and 2) gun homicides with female victims, but
not between gun ownership and overall rates of homicide and suicide. If our analysis shows that in
European countries higher levels of gun ownership are not systematically related with higher levels of
violent death, it follows that lower rates of gun ownership do not systematically translate in lower levels
of violent deaths. It also means that there is no negative correlation: higher levels of gun possession do
not translate into lower levels of violent deaths. In other words: there may be European countries with
high levels of gun ownership and low or average levels of violent death, but there also may be countries
with high levels of gun ownership and high levels of violent death.
This means that we need to look at other factors than gun possession to fully explain the variations in
levels of violent death in European countries; for example income inequality, unemployment rates,
poverty, mental illness, divorce rates and so on. One of the factors we must also take into account is the
potential impact of firearms legislation, a variable that we did not integrate in our correlational analysis.
Although we can expect a connection between the restrictiveness of firearms legislation and the
number of guns held by the population in a country, our analysis above tells us little about the possible
effects of stricter gun laws on the rates of violent deaths. The strictness of gun laws in European
countries may, however, help to explain why some countries have high levels of ownership but low or
average levels of violent death while other countries have equally high levels of gun prevalence and face
high levels of violent death.
Importantly, the question also remains open whether stricter guns regulations in Europe could be
instrumental in reducing the rate of homicides and suicides, especially in countries where these rates
FIREARMS AND VIOLENT DEATHS IN EUROPE P 45
are high. Could stricter legislation lead to fewer gun-related homicides and suicides? And could these
reductions also lead to a reduction in the overall levels of violent death, or will substitution of means
occur? In the next and final chapter we explore the question of the impact of stricter gun regulations on
rates of violent death in Europe, looking In particular at evidence from Austria, Belgium and Switzerland.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 46
5 The impact of stricter gun legislation?
The idea behind the introduction of stricter gun legislation is that it will reduce the risk of violent death
by influencing the availability of firearms. Limiting the legal ownership of firearms to persons with a
good cause (such as using them for hunting or target shooting), and to those who are unlikely to be a
danger to themselves or others (for example by excluding individuals with a history of violent behaviour
or severe mental problems), is assumed to reduce the misuse of legally-held firearms, while rules on the
storage of guns might prevent minors’ access to firearms and potential fatal outcomes resulting
therefrom. In addition, gun buy-back programmes may lower the general level of firearms ownership in
a country, while waiting periods
I
can provide a “cooling off” period which might avoid impulsive acts of
violence. Gun control proponents will argue that restricting the accessibility and availability of firearms
through stricter gun laws will lead to fewer gun-related deaths and therefore to decreased total rates of
violent death. Opponents of strict gun laws take issue with this thesis and argue that substitution of
means will take place: perpetrators will find other weapons to commit their acts. In this final chapter we
will test the hypothesis that the introduction of stricter gun legislation influences the gun-related and
overall rates of suicide and homicide in Europe.
Most studies on the possible effects of different aspects of firearms legislation and violent deaths are
focused on the situation in the United States. In general, the relationship between firearms legislation
and the number of firearms-related suicides is well-established. More discussion exists with regard to
the potential impact of firearms legislation on the overall suicide rates and on the prevalence of
firearms-related and overall homicide rates. In their review of the findings of - mainly American and
relatively old - cross-sectional studies on the relationship between firearms legislation and suicide, the
National Research Council observed that in general stricter gun laws have been associated with lower
rates of gun suicides, while the findings with regard to overall suicides rates have been less
consistent.
102
In 2005 the Task Force on Community Preventive Services reviewed a number of American
studies that focused on measuring the impact of firearms regulation (both federal and state laws) on
preventing violence. They concluded that because of the small number of available studies, limitations
in design and execution of these studies, and/or inconsistent or statistically insignificant effects, the
available evidence for each type of firearms regulation included in the study
II
was insufficient to
determine the effectiveness or ineffectiveness of firearms regulation on violent outcomes. They pointed
to the lack of a critical mass of high-quality studies evaluating the impact of gun legislation, and
I
This means that a specified number of days must pass between the moment of purchase of a firearm and the physical transfer
of the weapon to the purchaser.
II
The studies that were reviewed focused on one or a combination of seven types of regulation: bans on specific firearm sand
ammunition, restrictions on firearms acquisition, waiting periods, “shall issue” laws on concealed weapons carrying, child
access prevention laws requiring the safe storage of firearms, and zero tolerance of firearms in school.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 47
concluded that further research was needed to understand the potential impact of gun legislation on
the rate of violent deaths.
103
Most of these previous studies on the impact of firearms regulation on mortality have focused on
specific laws, not on the aggregate effect of all regulation. In a recent population-level study, Fleeger et
al analysed the relationship between an aggregate level of restrictiveness in firearms legislation
(measures to curb trafficking, strengthen background checks, ensure child safety, ban military style
assault weapons and restrict guns in public spaces) in all US states, on the one hand, and firearms-
related fatalities (overall, suicide and homicide) on the other hand, and controlled this relationship for a
number of socio-economic factors. Their analyses indicated that more restrictive gun legislation was
associated with lower rates of firearm-related fatalities, and little evidence of substitution was found.
Interestingly, they also controlled the relationship between the restrictiveness of firearms legislation
and firearms-related fatalities for household firearms rates, and observed that the relationship with
firearms-related suicides and homicides weakened. This suggests that part of the initially observed
relationship between firearms legislation and violent deaths can be explained by the rates of gun
ownership in these states. Of the different legislative categories, only background checks had a
significant correlation across all outcomes after controlling for firearm ownership rates.
104
Establishing causality between firearms regulation and the number of firearm -related and overall violent
deaths is a difficult endeavour, since these variables may affect each other and are not mutually
exclusive. While high-levels of violence may prompt policy action to increase the regulation of firearms,
an increased regulation of firearms might lead to a reduction in violence. Simple cross-sectional studies
alone, therefore, do not suffice for a thorough interpretation of the linkage since this interpretation
depends also on temporal sequence.
105
Besides population-level studies, the relationship between
firearms legislation and violent death has therefore also been researched using time-series studies in
which trends in mortality before and after the passage of a gun control law are analysed. The National
Research Council also reviewed the findings of a number of time-series studies, mainly conducted in the
1990s, and observed conflicting findings about the association between firearms legislation and
suicide.
106
An often-cited – non-American – example when discussing the possible effects of firearms legislation on
violent death is Australia. Several time-series studies have analysed the impact of the legislative reform
carried out in Australia in 1996 which was accompanied by a large-scale gun buy-back programme in
which more than 600,000 guns were collected - on firearms-related fatalities in the country. The results
of these studies, however, contradict each other. A number of studies show that the introduction of
more restrictive gun legislation across the country accelerated the declining trend in gun-related
fatalities, both suicides and homicides.
107
Importantly, Chapman et al also studied a possible
substitution effect. Since the overall suicide and homicide rates were relatively stable before the
FIREARMS AND VIOLENT DEATHS IN EUROPE P 48
legislative reform but declined afterwards, they concluded that there was no indication for a significant
substitution effect. Other studies using different methods, however, came to a different conclusion and
observed no significant effects of the new legislation on homicide rates.
108
In addition, the findings of
Baker and McPhedran suggested that the observed decrease in the number of firearms suicide was
accompanied by an increase in non-firearm suicide rates in the years following the law’s introduction.
However, these last studies have come in for much methodological criticism: on replicating the analyses,
and for example using longer time-series, Neill & Leigh found a statistically significant reduction in
firearm-related suicides and homicides and found no indication for a substantial method substitution
effect.
109
Their findings have, again, been questioned. Using different statistical Structural Break Tests,
Lee & Suardi came to the conclusion that there was little evidence to suggest that the new firearms
legislation had a significant effect on gun deaths.
110
The possible impact of gun legislation on violent deaths in Europe has received much less research
attention. In the following paragraphs we will analyze the results of three European time-series studies
on the impact of gun legislation on violent deaths, to see if the adoption of more restrictive firearms
legislation has an impact on the gun-related and overall rates of suicide and homicide. More precisely,
we will look at the rates of violent death from a temporal perspective and analyze trends in violent
deaths before and after the introduction of stricter firearms regulation in Austria, Belgium and
Switzerland. These cases are interesting for several reasons: first of all, because the introduction of new
legislation constituted a clear break in all three countries. Instead of incrementally implementing
changes in legislation, the Belgian government in 2006 drastically changed the regulatory framework.
This also happened in Austria when new firearms legislation was adopted in 1997 in order to comply
with the European Directive 91/477/EEC. Although the firearms legislation in Switzerland has not
changed drastically in recent years, and in a 2011 referendum Swiss voters rejected a proposal to
tighten controls on gun ownership
111
, a similar observation can be made vis-à-vis Switzerland since in
2003 reforms in the terms for military service severely restricted access to firearms across the country.
We can therefore clearly delineate the relevant policy interventions in terms of time. Secondly, since the
introduction of these regulations, a number of years have already passed, making it possible to collect
reliable and comprehensive data on trends in violent deaths since the regulation was implemented. At
the same time, the regulatory reforms took place not too long ago, so that comparable data can be
collected for the period before the introduction of the law. All these elements enable us to try and track
the effects of stricter gun laws. The main limitation of this time-series approach is that it is hard to
isolate the potential impact of the implementation of firearms legislation from other factors that
influence mortality, and which might also have changed over the same period. The analysis of the
Belgium case, which did not include control variables for possible confounders, is therefore exploratory
and the results should be interpreted cautiously. The analyses of the Austrian and Swiss cases, on the
other hand, have included such variables and can therefore be interpreted with more confidence.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 49
5.1 The impact of the 1997 firearms legislation in
Austria
In Austria a new more restrictive weapons act was adopted in 1997 (Waffengesetz 1996) following
the adoption of European Directive 91/477/EEC on the control of the acquisition and possession of
weapons. The new Austrian weapons act introduced the principle of a “good reason” for legally
possessing firearms, and imposed a number of additional restrictions on the ownership of certain
types of firearms including background checks, age requirements, psychological testing, safe
storage regulations and a cooling-off waiting period.
One of the first in-depth European time-series studies on the impact of firearms legislation on
violent death rates was made by Kapusta and colleagues in 2007, when they evaluated the impact of
the Austrian firearm legislation reform in 1997 on violent death rates as well as on the availability of
firearms in Austria by using the official statistics on suicides, gun homicides and firearms licences issued
from 1985 to 2005.
112
In their analysis, Kapusta and colleagues used Poisson regressions to compare
trends over time in gunshot suicides, in gunshot suicides as a percentage of total suicides, in overall
suicides and in gun homicides. For these variables they calculated time trends before 1998 and after
1998, and analyzed whether a statistically significant change in trend took place. In a second model,
unemployment rates and average alcohol consumption per capita were also included as covariates in
the analysis.
Interestingly significant changes in time trends after 1998 were found in the rate of gunshot suicides
(X2= 45.7, p<0.0001) and gun homicides (X2= 14.3, p<0.0001), even when controlled (and adjusted) for
unemployment and alcohol consumption rates (see table 10). While no significant time trend was found
in the gunshot suicide rate between 1985 and 1997, a significant negative trend was observed between
1998 and 2005. In this period the gunshot suicide rate declined by 4.7% annually
I
. A similar observation
was made with regard to gun homicides: while no significant trend over time was found before 1998, a
significant annual decrease of 2.3% per year was observed after 1998. These results clearly suggest that
the introduction of more restrictive firearms legislation effectively reduced the rates of firearm suicide
and homicide.
The authors also investigated whether there was a significant change in the time series relating to
overall suicide rates. While the overall suicide rates decreased from 27.6 per 100,000 persons in 1985 to
I
Kapusta et al also analyzed the firearms suicide rates by age group and gender. A significant trend change was observed for
women aged 20-64 years, men aged 20-64 years and men aged 65 years or older. For younger persons and women aged 65
years or older, no significant change in time trend was observed.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 50
16.7 per 100,000 persons in 2005, no significant change in time trend was observed after the
introduction of the new legislation. This means that the decrease in overall suicide rates that was
observed before 1998 did not accelerate significantly afterwards. At first sight this suggests that a
substitution of methods in suicides took place after the introduction of the new legislation. However,
although not included in their tables, Kapusta and colleagues also analyzed the rate of suicide by other
means than firearms. Since no significant increase was observed here, they concluded that their
analyses did not suggest the existence of a substantial substitution effect. The authors added that after
the introduction of the more stringent firearm law, the numbers of firearms licences issued decreased.
They concluded that the decline in firearm-related deaths seemed to have been mediated by the legal
restriction of firearm availability. No analyses were made of overall homicide rates.
113
Table 10: Impact of firearms legislation on suicide and homicide rates (1985-1997 and 1998-2005) in
Austria: parameter estimates derived from the Poisson regression model
Growth factor
(95% CI)
Change in growth
factor % (95% CI)
Test for trend
change
Before 1998
After 1998
X2
P
Model 1 (unadjusted)
% of firearm suicides among all
suicides
1.026
(1.019 to 1.032)
0.971
(0.961 to 0.981)
-5.2
(-6.7 to -3.9)
53.0
<0.0001
Firearm suicide rate
1.000
(0.994 to 1.010)
0.949
(0.935 to 0.964)
-4.9
(-5.9 to -3.9)
45.7
<0.0001
Total suicide rate
0.975
(0.971 to 0.979)
0.978
(0.971 to 0.985)
0.3
(-1.0 to 1.3)
0.28
0.59
Firearm homicide rate
1.002
(0.977 to 1.029)
0.877
(0.832 to 0.923)
-12.5
(-18.2 to -6.2)
14.3
<0.0001
Model 2 (adjusted)
% of firearm suicides among all
suicides
1.026
(1.016 to 1.038)
0.978
(0.963 to 0.993)
-4.8
(-6.7 to -2.8)
21.8
<0.0001
Firearm suicide rate
1.00
(0.968 to 1.013)
0.953 (0.937 to
0.969)
-4.8
(-6.9 to -2.7)
19.9
<0.0001
Total suicide rate
0.975
(0.969 to 0.982)
0.974
(0.964 to 0.984)
-0.1
(-1.4 to 1.3)
0.03
0.87
Firearm homicide rate
0.994
(0.894 to 1.10)
0.895
(0.791 to 1.013)
-9.9
(-18.9 to -0.1)
3.9
<0.0001
Source: Kapusta et al (2007)
114
FIREARMS AND VIOLENT DEATHS IN EUROPE P 51
5.2 The impact of the 2006 weapons act in Belgium
In Belgium, a stricter weapons law was enacted in June 2006. The new legislation prescribed that all
firearms required a licence unless they were either categorized as ‘prohibited’ or as ‘freely obtainable’
weapons (the latter category was drastically reduced compared with the previous control regime). As
prescribed by the European firearms directive, the new law implemented the principle of the ‘good
reason’, limiting civilian gun ownership to those who could prove they needed their guns for specific
purposes. The law of 2006 also listed a series of personal criteria that had to be fulfilled before an
authorization could be obtained, such as not having been convicted for certain crimes, not having been
interned as a mentally ill person, presenting a medical certificate proving mental and physical fitness,
and succeeding in theoretical and practical tests. In addition, local police forces were to assess all
applications for licences, and adults living with the applicant had to attest that they approved of having
a gun in their household. Other stipulations in the Belgian weapons act concerned rules for the sale and
purchase, the marking and registering, and the safe storage and transportation of firearms. In addition
to the new provisions, a collection and amnesty programme was set up in Belgium after the introduction
of the new law in which more than 200,000 firearms were collected and later destroyed. Assessing the
exact impact of the collection programme, however, is difficult since the authorities responsible did not
use uniform registration methods. It is therefore unclear what types of firearms were handed in and
whether they were legally or illegally held.
115
In a recent study exploring the effects of the introduction of the 2006 weapons act in Belgium, Duquet &
Van Alstein analyzed trends in the gun-related and overall suicide and homicide rates. To track violent
death rates before and after the introduction of the new law (2003-2009) they relied on data from the
European Detailed Mortality Database of the WHO Regional Office for Europe. The figures for Belgium
show that in the period 2003-2010 more than 2,000 persons died as a result of firearm injuries. These
figures show a strong decrease in the number of gun deaths in the years following the change of
legislation in 2006: the annual average of gun-deaths in 2007-2010 decreased by 35% compared with
the period 2003-2005 (see table 11). No regression analyses were performed on this dataset.
As in Europe in general, most of the firearms-related deaths in Belgium are the result of successful
suicide attempts. Between 2003 and 2010, 1,593 deaths were classified as firearms-related suicides,
which comes down to an average of almost 200 gun suicides per year in Belgium. The data indicate a
significant decrease (-34%) in the number of gun suicides in the years following the change of
legislation: from an annual average of 245 gun suicides in 2003-2005 to 169 gun suicides in 2007-2009
I
.
Importantly, the overall number of suicides also decreased (-4%) between these periods, and it appears
that almost the entire decrease in total suicides can be attributed to the decrease in gun suicides after
2006 (see figure 11).
I
Since the legislation changed in mid-2006, we do not take the figures for 2006 into account but only the years before and
after the change of legislation.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 52
Table 11: Firearms-related deaths, Belgium, 2003-2010
Total
mortality
with firearm*
Suicide
Homicide
With firearm
total
With firearm
Total
2003
330
261
2,091
53
144
2004
309
234
1,986
59
178
2005
315
240
2,028
46
163
2006
256
207
1,934
31
146
2007
210
165
1,856
35
157
2008
231
189
2,000
29
128
2009
189
153
2,013
33
134
2010
197
144
2,011
36
117
Total
2,037
1,593
15,919
322
1,167
Annual average 2003-2005
318.0
245.0
2,035.0
52.7
161.7
Annual average 2007-2010
206.8
162.8
1,970.0
33.3
134.0
* The total mortality rate involving a firearm includes, besides suicides and homicides, also a small number of
accidents (average: 2,4 per year) and deaths in which the cause is unclear (average: 13,7 per year). These statistics
are not included in table 11 but can be obtained from the authors.
Source: Duquet & Van Alstein (2015)
116
Figure 11: Annual average of firearms-related and other suicides, Belgium, 2003-2005 and 2007-2010
Source: Duquet & Van Alstein (2015)
117
The number of firearms related homicides also dropped significantly (-39%): from 52.7 annual deaths in
2003-2005 to 33.3 deaths in 2007-2010. Further, the overall number of homicides decreased between
these periods (-17%). As with the suicide data, almost the whole decrease in total homicide figures can
be attributed to the decrease in gun-related homicides (see figure 12).
1,790 1,807
245 163
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2003-2005 2007-2010
gun suicides
other suicides
FIREARMS AND VIOLENT DEATHS IN EUROPE P 53
Figure 12: Annual average of firearms-related and other homicides, Belgium, 2003-2005 and 2007-2010
Source: Duquet & Van Alstein (2015)
118
These analyses indicate that the implementation of a more restrictive firearms legislation in Belgium has
led to less gun-related deaths, but importantly also suggests a decrease in the overall suicide rates, and
possibly also in the overall homicide rates. While very marked decreases in gunshot suicides and gun
homicides can be observed, the number of suicides and homicides committed with other means than
firearms largely remained at the same level. This suggest that, although some people who did not have
access to firearms chose a different method to commit suicide or homicide, most did not, and taken
together no substantial substitution of means took place.
This analysis of the Belgian case is exploratory and has clear limitations. First of all, the available time
series is short.
I
Secondly, the statistical analysis used is rather crude: for example, no interrupted time
series analysis to control for pre-existing temporal trends was used
II
. Thirdly, no controls for relevant
socio-demographic variables such as unemployment, rates of mental disorders or substance
dependence were included. This makes it difficult to conclusively link the decrease in the total number
of suicides to the introduction of the weapons law: other factors and societal trends could also have
played a role. Fourthly, as the number of homicides under study was small, it is difficult to draw strong
conclusions regarding a decrease in the homicide rate.
I
In their report Duquet & Van Alstein (2015) also analyzed the available mortality data for the Flemish Region (the Northern
part of Belgium). Although the population in this region is smaller, the time series that are available were longer (2000-2012).
Similar results were found for the Flemish Region as for Belgium as a whole.
II
This technique uses data from the pre-intervention to compute a forecast for the post-intervention period, which is then
compared with observed post-intervention data.
109.0 100.8
52.7
33.3
0
20
40
60
80
100
120
140
160
180
2003-2005 2007-2010
gun homicides
other homicides
FIREARMS AND VIOLENT DEATHS IN EUROPE P 54
5.3 The impact of military service reforms in
Switzerland in 2003
In 2013 Reisch and colleagues studied the change in suicide rates in Switzerland before and after the
firearms restriction resulting from reforms in the military system in 2003 (1995-2008). Switzerland
historically uses a militia army for its defence. This means that during their time of service, many men
keep army-issued guns in their homes throughout the year. After their service, soldiers can buy these
service weapons for a small fee. Most soldiers and veterans retain this single military gun but do not
possess other firearms. At the same time, the authors note, Swiss firearms laws are generally less
restrictive compared with other European countries .
A significant share of the gunshot suicides in Switzerland are committed with army weapons. An analysis
of suicides committed in 2004 indicates that in 44% of all gunshot suicides an army weapon was used.
119
Given the association of the large number of gunshot suicides with army weapons in Switzerland, and
the high lethality of these weapons, changes in the regulation of these firearms have been identified as
key elements for successful suicide prevention programmes.
In 2003 the Swiss army was restructured, reducing troop numbers from approximately 400,000 to
200,000. This had a significant impact on the availability of military guns. Moreover, the conditions for
buying these guns after the end of service were made more restrictive by increasing the fee and the
requirement to take out a licence. In their study, Reisch and colleagues analyzed the impact of the
military reform on suicide rates by using an interrupted time-series approach in which detailed data
from a pre-intervention period were used to compute a forecast for the post-intervention period. Their
analysis was based on monthly data on suicides, based on death certificates provided by the Federal
Statistical Office. Possible confounders such as a number of socio-demographic variables, and various
other variables such as levels of mental disorders, were included in the statistical models. In a second
step, the forecast values were compared to the observed values after 2003. If statistically significant
differences were seen between the forecast and observed values, there was a clear indication of an
impact of the newly introduced legislation. The authors tested the validity of their approach by using
several alternative statistical approaches.
The analyses of Reisch et al. showed a reduction both in the overall suicide rate and in the firearms
suicide rate in the case group (men aged 18-43) in a five-year period after the military reform of 2003
(2003-2008). No significant increases were found for suicides committed with other means than
firearms, although an increase in railway suicides was observed. In addition, the comparison groups
(women aged 18-44 and men aged 44-53) showed no statistically significant changes in their firearms
suicide rates and overall suicide rates. These results suggest a clear impact of the introduction of the
new legislation on gunshot suicides and overall suicide rates. Interestingly, a partial method-substitution
FIREARMS AND VIOLENT DEATHS IN EUROPE P 55
effect was found. Using their more advanced statistical approach, the authors were able to calculate this
substitution effect and estimated that 22% of the reduction in firearms was substituted by other suicide
methods. While some persons used a different way of committing suicide when access to firearms was
no longer possible (especially railway suicides), more than three quarters of the risk group prone to
gunshot suicide did not switch to other means. Reisch and colleagues therefore conclude that the
restriction of firearm availability in Switzerland resulting from military reforms was followed by an
enduring decrease in the general suicide rate.
120
Figure 13: Suicide rates in Switzerland before and after implementation of the 2003 Army XXI Reform
I
*** p<0.001. n.s.=not significant
Source: Reisch et al. (2013)
121
I
The Army XXI reform was implemented during the period from March 1, 2003, to the end of February 2004. Each depicted
year reflects the overall suicide rate for a 12-month period starting in March and ending in February the following year. In
2008, data was available up to December 31 and the suicide rate was adjusted accordingly.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 56
Conclusion
Many studies have been carried out in recent decades on the linkages between violent deaths, gun
ownership and/or firearms legislation. Yet only a rather small selection of these studies has focused on
the situation in Europe, partially because gun death rates in Europe are traditionally rather low by global
standards. While European policy-makers are paying increasing attention to gun violence, there is a lack
of reliable and comprehensive information on the linkages between gun ownership and legislation on
the one hand, and gun violence on the other hand, in Europe and this in turn impedes the development
of appropriate and effective firearms policies. In this report we have therefore explored the relevant
linkages for Europe, focusing on lethal gun violence and the connections between this violence, gun
ownership and firearms legislation. Setting out into largely unexplored territory, we took with us the
lessons of the few earlier studies on Europe and, in particular, the more developed body of evidence
from other parts of the world.
In the first chapter we analysed the available data on gun ownership in Europe and discussed the
methodological limitations of this data. In the second chapter we focused on gun-related deaths in
Europe and explored the data on gun-related suicides and homicides. In the following chapters we
focused on the findings of previous studies on the linkage between gun ownership and violent deaths
(chapter three), and performed a new population-level analysis specifically for European countries
(chapter four). In chapter five we reviewed the literature on the impact of firearms legislation on violent
death rates, focusing particularly on the results of three recent European time-series studies analysing
this impact. Before coming to our main conclusions on the linkages between gun ownership, firearms
legislation and violent deaths in Europe, we will first give a brief overview of our main findings both on
firearm possession and on gun-related deaths in Europe.
Gun ownership
At the moment there are at least 25 million gun owners, mainly men, in the European Union (EU), who -
according to estimates by the Small Arms Survey - possess some 79.8 million firearms. This corresponds
to around 15.7 firearms per 100 persons. There seems to be no general trend of increasing or
decreasing ownership in European countries, but the limited available longitudinal data suggest that
there seems to be a decrease in these rates, especially in countries with traditional high rates of gun
ownership. Hunting is the reason most often cited for firearms possession in the EU, followed by
professional reasons and sports. Gun possession is less frequently motivated by personal protection and
collection purposes. Given the importance of hunting, it is no surprise that gun owners in most
European countries mainly possess shotguns and/or rifles. The possession of handguns is not equally
widespread in Europe. An interesting observation is that the motivations for gun possession can differ
FIREARMS AND VIOLENT DEATHS IN EUROPE P 57
greatly between countries: in the older EU Member States citizens are generally more likely to own
firearms for hunting and sports, while in the more recent Member States gun ownership is more often
linked to professional reasons and personal self-defence. The available data also suggest that owning
guns for professional reasons or personal protection was probably more prevalent before than it is
nowadays.
Given the lack of reliable and comparable official figures on firearms possession in most European
countries, the above-mentioned figures are based on the results of self-reporting surveys and expert
estimates. Although the figures on gun possession used in this report draw on the most reliable
currently available estimates for firearms possession in Europe, they need to be interpreted with
caution. More official figures will become available in the near future, since Directive 2008/51/EC
(amending the Firearms Directive 91/477/EEC) stipulates that all EU MS must establish and maintain a
computerized data filing system on civilian-owned firearms. This will strongly enhance the authorities’
prospects of knowing the numbers and types of legally held firearms present in their country. The total
number of illegally-held firearms in European societies is harder to estimate. There is an urgent need for
solid research into the illicit firearms market in Europe.
Gun-related violent death
In the EU over recent years, an average of approximately 6,700 persons have died annually as a result of
gunshot wounds. Two countries aside, a strong downward trend in the number of gun deaths can be
observed across Europe. Between 2000 and 2012 the number of annual gun deaths decreased by 19% in
Europe overall, with an especially spectacular decrease in gun deaths in Central and Eastern Europe.
Most gun deaths (75%) in Europe are the result of successful suicide attempts. In the EU, 5,000 persons
commit suicide every year using a firearm. Approximately 9% of all successful suicide attempts in Europe
in 2000-12 were undertaken with a firearm. The overwhelming majority of gunshot suicide victims (96%)
were men, particularly older men. Women and minors make up only a very small proportion of all gun
suicides in Europe. Our analyses suggest that in the Nordic and Western European countries, long guns
are used in most suicides, while in the Eastern European countries handguns are more frequently used.
The distribution of types of firearms used for committing suicides probably reflects the general (legal)
ownership rates of the different types of firearms in these countries. Reliable data on the legal status of
the guns used are lacking in the WHO database. The results of most previous studies, however, indicate
that the majority of gun suicides are committed with legally-held firearms.
Homicides make up 15% of all firearms-related deaths in Europe. On an annual basis, an average of
1,000 homicides are committed with a firearm in the EU, but significant national differences in gun
homicide rates can be observed. In approximately 20% of all homicides in Europe a gun was used, which
is considerably lower than the global average (41%). Despite the large national differences mentioned,
FIREARMS AND VIOLENT DEATHS IN EUROPE P 58
what all European countries have in common is that men are much more likely to become a gun
homicide victim than women. The available data further suggest that handguns are used more
frequently for homicides than for suicides. Interestingly, most female homicide victims are older women
while younger men in particular have a higher gun homicide rate. This can probably be connected to the
different types of contexts in which homicides take place: while women are typically killed at home by
their (ex-)partner or in a similar context of interpersonal violence, men are killed in a wider range of
settings, including the context of organized crime or as a result of financial disputes. Our analyses have
found a general downward trend in the number of gun homicides across Europe, and an especially
spectacular decline can be observed in Eastern European countries and the Baltic States. Information is
lacking on the legal status of the firearms used to commit these homicides.
Linkages between gun ownership, firearms legislation and violent deaths
Our population-level analysis of the relationship between gun ownership rates and violent death rates in
Europe indicates a strong positive correlation between gun ownership rates in a country and the rate of
firearms-related deaths, especially gunshot suicides with male victims. With regard to gun homicides a
statistically significant correlation was found only for homicides with female victims. In other words, in
European countries with lower rates of gun ownership we will typically find fewer gun deaths, fewer
men committing suicide with a gun, and fewer women being killed with a firearm. These findings are
largely in line with previous studies on the situation in the United States and with cross-national
comparisons. Most opponents of stricter gun laws have acknowledged this observation, but argue that if
people are really determined to commit suicide or homicide they will find a way to do so, regardless of
whether firearms are readily available. We concur with this statement: but it needs to be nuanced and,
more importantly, it does not imply there is no room for policy to intervene to prevent suicide and
homicide.
As shown in this report, a significant proportion of all suicides are rather impulsive acts while a
significant share of homicides are the result of expressive violence, especially in the domestic sphere.
Previous research suggests that in these contexts in particular, firearms are a frequently used means of
committing violence. Studies further indicate that a significant proportion of these firearms are legally
held: a majority of suicides and a portion of the homicides are committed with legally-held firearms, for
example with service weapons. This observation provides us with potential for effective policy
intervention. Especially in the context of impulsive acts, limiting access to firearms can not only play an
important role in delaying suicide or homicide attempts, but also in impeding possible future attempts.
The goal of firearms legislation in Europe is to restrict access to firearms to persons who are capable of
using and storing a firearm in a responsible manner. Several restrictions such as the need to cite a good
reason for possessing firearms, background checks, and safe storage rules are therefore typically
included in European firearms regulations. Other measures that can be taken to limit the misuse of
firearms include prohibiting the storing of service weapons at home. The Swiss example in chapter 5
FIREARMS AND VIOLENT DEATHS IN EUROPE P 59
clearly indicates that this last example can have a strong impact, not only on gun-related suicides, but
also on overall suicide rates.
Obviously, a substitution of means can occur in the context of suicide or homicide attempts, but the
analyses in this report indicate that there is not a full but rather a partial substitution effect. Our
population-level analysis of the relationship between gun ownership rates and suicide and homicide
rates in Europe shows that the low gun suicide and homicide rates typically found in countries with low
levels of gun ownership are not compensated by significant higher rates for suicide and homicide
committed by means other than a firearm. Similar results were found in the time-series studies that we
analysed on the impact of firearms legislation: in neither of these studies was the strong observed
decrease in gun deaths in the years following the introduction of new legislation accompanied by a
significant increase in the rates of suicide and homicide committed with other means than firearms. In
other words, some people will find a different means of committing suicide or homicide if they do not
have access to firearms, but others will not do so. In addition, the highly lethal nature of firearms as an
instrument for suicide or homicide implies that even if some degree of substitution occurs, this will
typically involve less lethal means, which in turn should lower the success rate of these attempts and
thus prevent some violent deaths. These findings strongly suggest that implementing firearms
regulation that curbs access of firearms has a strong impact not only on the number of gun deaths in a
given country, but also in lowering the overall levels of violent death.
The way ahead
The results of the time-series studies on Austria, Belgium and Switzerland presented in this report
illustrate the impact of firearms legislation on gun suicides, and possibly also on homicides, which in
turn affects overall suicide and homicide rates. What is unclear, however, is which specific element of
the firearms legislation has the greatest effect on violent death rates: the principle of citing a good
reason for possession, background checks, waiting periods, safe storage rules, or the amnesty and gun
collection programmes that usually follow the introduction of new legislation. This is a very important,
but difficult question that requires more in-depth research to provide a conclusive answer. Even in the
United States, where much more research has been undertaken on this topic, a review of previous
studies suggests that the available evidence so far is inadequate for reaching solid conclusions. It is thus
imperative to set up more research on the possible impact of gun legislation on gun violence in different
European countries: research that should also focus on non-lethal gun violence, and address issues such
as the legal status of the firearms used and the consequences of this gun violence for society.
As we have argued consistently throughout this report, more research is needed into European aspects
of gun control. It is also important in this context that European governments increase their efforts in
generating and managing appropriate, comprehensive and solid statistical information on the levels of
firearms ownership, violent incidents and gun-related crime. Solid statistical information constitutes the
FIREARMS AND VIOLENT DEATHS IN EUROPE P 60
basis for further scientific research. To be useful, these statistics need to be adequately detailed. Gun-
related crime, for example, is not a single, unified phenomenon. It has many aspects and covers many
different offences ranging from illegal possession and trafficking to armed robbery and murder with a
gun. Crime statistics should distinguish between these different aspects of gun crime in order to do
justice to its complexity. In other words, good statistics are a necessary condition for addressing gaps in
our knowledge of gun laws and firearms in European societies. Anything that governments can do to
generate and manage more data on firearms will, of course, also serve more immediate policy
objectives: effective investigative and prosecutorial capabilities and strategies also depend on good
intelligence and a good, realistic picture of gun-related crime. With the present report, we hope to have
contributed to this project of generating more empirical knowledge about gun control in the European
context.
FIREARMS AND VIOLENT DEATHS IN EUROPE P 61
Annex
Table 12: reasons for firearms possession by respondents who used to own a firearm, per country,
2013
Hunting
Sports
Collector
Professional
reasons
Personal
Protection
Other
personal
reasons
Number of
respondents
Austria
15%
20%
15%
33%
28%
0%
46
Belgium
19%
18%
16%
20%
37%
12%
74
Bulgaria
10%
6%
2%
65%
14%
3%
63
Croatia
16%
6%
6%
49%
16%
10%
110
Cyprus
21%
6%
0%
73%
2%
2%
67
Czech Republic
18%
28%
6%
51%
29%
8%
51
Denmark
37%
19%
5%
17%
4%
14%
79
Estonia
25%
16%
2%
38%
22%
0%
64
Finland
58%
23%
3%
8%
5%
13%
40
France
39%
18%
3%
21%
11%
16%
71
Germany
6%
32%
7%
24%
9%
25%
68
Greece
56%
8%
3%
31%
18%
8%
39
Hungary
29%
10%
0%
56%
15%
0%
41
Ireland
62%
25%
1%
12%
5%
3%
81
Italy
26%
10%
7%
45%
13%
7%
31
Latvia
20%
3%
5%
56%
14%
5%
59
Lithuania
13%
13%
2%
37%
32%
7%
62
Luxembourg
21%
29%
21%
42%
8%
17%
24
Malta
46%
18%
0%
36%
9%
9%
11
Netherlands
7%
29%
2%
60%
7%
5%
42
Poland
8%
11%
0%
62%
19%
4%
53
Portugal
21%
8%
3%
55%
19%
15%
73
Romania
5%
3%
2%
78%
10%
0%
58
Slovakia
11%
19%
11%
44%
19%
0%
36
Slovenia
15%
29%
0%
37%
4%
10%
52
Spain
58%
28%
0%
17%
6%
8%
36
Sweden
37%
20%
6%
22%
0%
18%
49
United Kingdom
23%
45%
5%
33%
20%
10%
40
Source: Eurobarometer
122
- own calculations
FIREARMS AND VIOLENT DEATHS IN EUROPE P 62
Table 13: Firearms-related suicides (also as % of total suicides) in 33 European countries, 2000-2012
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Average
/year
Age-
standardized
gun suicide
rate per 100
000I
Suicides as % of
total gun
deathsII
Gun suicides
as % of all
suicidesIII
Austria
261
227
215
233
228
195
196
204
225
205
217
218,7
1.96
94.3%
17%
Belgium
261
234
240
207
165
189
153
144
199,1
1.11
73,1%
7%
Bulgaria
55
75
57
71