Compliance with the European Union common position on arms exports : a case study on arms exports to Saudi Arabia

(1)

Universiteit van Amsterdam

Master in Sociology

Social Problems and Social Policy Track

Master Thesis

Robin Hübscher

robinhuebscher@bluewin.ch

Student nr.: 11125616

Supervision:

First supervisor: prof. dr. Jan Willem Duyvendak

Second supervisor: drs. Paul Aarts

Amsterdam, 30

th

_{June 2016}

A Case Study on Arms Exports to Saudi Arabia

(2)

ii

Abstract

Saudi Arabia became the biggest arms importer in 2015 and EU member states contributed substantially to the Saudi imports. With a Common Position on arms exports, all EU member states are legally bound to `responsible` arms export controls. It is argued that states commit to responsible arms export controls in order to improve the international reputation. With poor accountability mechanisms it is mainly arms trade scandals which push states from commitment towards compliance. With a case study on EU arms exports to Saudi Arabia it is explored whether the Common Position was applied consistently by the EU member states. The case study shows that virtually all arms export licences were granted. Denied licence applications can partly be linked to arms trade scandals and point to concerns on human rights violations which, however, did not prevent exports worth billions from being issued. The results support the assumption that states commit to responsible arms export controls mainly to improve the international reputation which questions the preventive element of arms export controls.

(3)

iii

List of Abbreviations

APC Armoured Personnel Carrier ATD Arms Transfer Database ATT Arms Trade Treaty

CCW Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons which may be Deemed to be Excessively Injurious or to have Indiscriminate Effects

CFE Treaty on Concentional Armed Forces in Europe COARM Working Party on Conventional Arms Exports CoC Code of Conduct

CP Common Position HRW Human Rights Watch

ICCPR International Covenant on Civil and Political Rights MED Military Expenditure Database

Miksa Ministry of Interior Kingdom of Saudi Arabia ML EU Common Military List

NPE Normative Power Europe

PoA Programme of Action to Prevent, Combat and Eradicate the Illicit Trade in Small Arms and Light Weapons in All Its Aspects

SALW Small Arms and Light Weapons TIV Trend Indicator Value

UDHR Universal Declaration of Human Rights UNROC United Nations Register of Conventional Arms WMD Weapons of Mass Destruction

(5)

1

1 Introduction

Saudi Arabia became the biggest arms importer in 2015 with arms imports worth $3.2 billion. The military expenditure of Saudi Arabia is the third highest worldwide and, measured as a share of GDP, Saudi Arabia devotes seven times more financial resources to the military than the worldwide average. Arms exports from EU member countries account for 42% of the total arms imports of Saudi Arabia and have thus contributed substantially to the huge increase of Saudi arms imports since the turn of the millennium. A sharp contrast to the flourishing arms trade between EU countries and Saudi Arabia is the EU parliamentary vote in favour of an arms embargo against Saudi Arabia in February 2016 (EU Parliament: RSP 2515). The call for an arms embargo comes after allegations of serious breaches of international humanitarian law by Saudi Arabia in the Yemeni war which started in early 2015, the same year in which arms exports to Saudi Arabia rose to a new high. Arms exports from the EU are subject to transfer controls, which are an important preventive element of arms controls (Bauer 2010: 311). Arms transfers are the responsibility of the member states but a key common denominator is the legally binding “Council Common Position 2008/944/CFSP of 8 December 2008 defining common rules governing control of exports of military technology and equipment” (EU Council 2008). Before becoming legally binding, the Common Position (CP) was a Code of Conduct (CoC) which was introduced in 1998. The CP is aimed at strengthening and harmonizing the export policies of member states (Bromley 2008) by setting minimum standards in the form of eight consolidated criteria against which exports need to be assessed.

Several studies and NGO reports on the CP cast doubt on the success of harmonizing EU countries arms export policies and point to an inconsistent application. Bromley and Broszka (2008) assess the impact of the CoC ten years after it was introduced. They find little evidence for harmonization of EU member states arms exports but observe an increased restrictiveness regarding arms exports under the human rights and conflict criteria of the CoC. Erickson (2011) looks at EU arms exports in connection with human rights, conflict and democracy in the recipient country and conducts a case study on the China embargo. She concludes that domestic material and normative interests remain important in explaining member states arms exports. The events of the Arab Spring have further generated a lot of attention regarding arms exports from EU member states to countries which were affected (Bromley 2012; Duquet 2014; Hansen/Marsh 2014). Findings on arms exports prior to the Arab Spring point to a lax and inconsistent application of the CoC and CP.

This case study on EU arms exports to Saudi Arabia is motivated by the concurrence of the above: The relevance of Saudi Arabia as biggest arms importer, the controversial nature of arms exports to Saudi Arabia exemplified by the EU parliamentary vote in favour of an arms embargo and study results pointing to an insufficient or inconsistent application of the CP. The main research question addressed with the case study is as follows:

1. Is the Common Position applied consistently by the European Union member states regarding the control of arms exports to Saudi Arabia?

(6)

2

Looking at the consistency of the application of the CP, notifications on denied licences for arms exports as reported in annual EU Council reports, are an important information source. These reports include information on granted and denied licences broken down by categories of military equipment as laid out by the EU Common Military List (EU Council 2015). Regarding licence denials, there is also noted on which of the consolidated criteria the denial is based upon. The information on denied equipment and cited criteria can give some insight into what could be the reason for the denial which leads to the following two sub-questions:

2a. Licences for which Military List categories were denied? 2b. What are the most reported criteria leading to licence denials?

Although there have been improvements in transparency, public information on the arms trade is scarce, especially when it comes to the reasoning behind export decision. Nonetheless should possible explanations of licence decisions not be left out in order to go beyond the descriptive character of the first research questions. An explanation is sought with the third sub-question:

3. What could explain differing applications of the Common Position?

This is arguably the most interesting part of the topic but it is also very tricky, not to say impossible, to explain what lies behind the decisions whether or not to grant a licence. This is reflected in the use of the word could in the question. Any answer of this question will be, to a bigger or smaller extent, speculative. Explaining inconsistent applications of the CP relies on assumptions derived from the theoretical notion of reputation. Erickson (2015a) argues, that it is reputation which pushes states towards commitment to stricter arms controls but compliance to these commitments comes from domestic pressures. Reputation thus helps explaining the often ascertained gap between commitment and compliance. The starting point of Erickson`s analysis is the UN Arms Trade Treaty (ATT) which entered into force in 2014 (UNODA: ATT). The ATT has many similarities to the CP but is in scope much broader with 130 states having signed the treaty so far. It is too early to assess the impact of the ATT but findings on the application of the CP also have relevance for the ATT and the theoretical implications are not bound to a specific treaty.

The paper is structured as follows. The theoretical notion of reputation is discussed in chapter two, which also addresses the question why states sell arms in the first place. The third chapter provides an overview of the developments of arms controls and key elements of the CP. Chapter four outlines the method and data used to address the research questions. In chapter five arms procurement of Saudi Arabia and the arms trade between EU member states and Saudi Arabia is discussed. The core of the case study are the chapters six and seven. While chapter six is devoted to the implementation of the CP regarding arms exports to Saudi Arabia, chapter seven provides a discussion of the research questions in light of the results of the case study. The case study is rounded off with a conclusion in chapter eight.

(7)

3

2 Arms Trade and Reputation

This chapter consists of a theoretical discussion of the arms trade and arms transfer control. The first part explores the reasons why states sell arms to other countries in the first place. The second part is aimed at explaining why states agree to arms transfer controls. An obstacle to explain this commitment is that compliance to arms transfer controls is often not a given. This gap between commitment and compliance is acknowledged here with the theoretical notion of reputation.

2.1 Why Do States Sell Arms?

While there are several reasons why states have a demand for arms there are basically two factors which explain why states sell arms: Political and economic (Wezeman 2010). Arms supplies were used during the Cold War as an incentive to gain allies and were necessary for these allies to fight off external or internal enemies. This political reason to sell arms has not disappeared after the Cold War. Especially the USA reward US-friendly policies by sending arms to other countries as aid or to preferential conditions. With the war on terror after the 9/11 terrorist attacks, the political factor in arms supplies has again become more important for the USA (Wezeman 2010: 199).

More important than the political factors, Wezeman (2010: 200) argues, are the economic reasons. For the state, arms exports have a direct impact on employment, the balance of trade and government income through taxation (ibid.). For the arms producing companies, exports can lower the price per unit by spreading the costs of research and development over more units. Even more important than profit or jobs might be the maintenance of a defence industrial base. Development and production of arms has become more expensive and without exports, some countries would need to close down parts of their arms industry (Wezeman 2010: 200). However, closing down parts of the arms industry means more dependence on other countries.

The end of the Cold War triggered also a shift from political reasons for arms exports to more economic reasons. The demand for arms decreased after the Cold War because of a reduced-threat security environment and there was a surplus of weapons from former Eastern-Bloc countries (Erickson 2011). The size of the armed forces was cut down in many countries by over 50% with a corresponding drastic reduction of inventory and lower domestic demand of arms (Wezeman 2010: 205). Trying to avoid domestic political penalties from job losses, states started using arms exports to keep the defence industrial base healthy. Many jobs were nonetheless made redundant worldwide because of lower production numbers (Bitzinger 2010). Factories had to close down and arms production got more concentrated which led to the emergence of several `mega-defence firms` (ibid.).

Although there is again an increase in arms transfers since around 2002 (Wezeman/Wezeman 2014), the global arms market has become more complex and competitive (Bitzinger 2010). It is often described as a buyer`s market because arms exports developed from a supplementary income to a critical part of the defence industrial base (ibid.). To achieve an arms deal, exporters commonly offer incentives to potential

(8)

4

buyers in the form of offsets (industrial participation), technology transfer and foreign direct investment (Bitzinger 2010: 209). Arms exporting countries can also have economic motives which go beyond the direct arms sale. Such a motive can be the access to markets for non-military products through the supply of arms as a sign of friendliness (Wezeman 2010: 200). Another motive can be raw materials such as oil. The USA supplied arms to oil-producing states in the 1970s to access these energy resources but also more recent arms transfers to the Middle East and Africa are suspected to have been motived by the same reason (ibid.).

In Europe, previously state-owned arms companies were privatized after the Cold War, most radically in the UK while other states still own a big part of the industry (Dunne/Sköns 2010). The connections between states and companies are, however, not always that clear and differ between countries (ibid.). Although there have been mergers and increased international cooperation within Europe, the national basis for arms production is still predominant and key technologies are protected nationally (Bitzinger 2010: 212). The relationship between states and companies changed since the end of the Cold War with a reduction of direct links (Dunne/Sköns 2010). National Governments play nonetheless still an important role and indirect links between states and companies remain powerful, although less visible (ibid.).

Many of the reasons, political and economic, are brought forward to support arms transfers in the public. These reasons are, however, not undisputed and political and economic gains have been described as exaggerated. Mayhew (2005) looks at the arguments given by the UK labour government in support of arms exports. Regarding the economic factor, Mayhew argues that arms exports make up only a small part of the balance of trade (1.6% in the UK) and that the job-argument is unsustainable because only 0.24% of all jobs in the UK are sustained by arms exports. A cut in military exports does further not mean a definite loss of jobs because other jobs in non-military sectors will be created (Chalmers et al. 2002). Also regarding the political and strategical arguments, Mayhew finds little support.

Regardless of these reservations about the benefits of arms exports, billions worth of arms are exported every year and the importance of arms exports is stressed by politicians all over the world. The question then arises, why states agree to arms transfer control agreements involving mainly costs instead of benefits? This question will be explored in the next chapter.

2.2 Why Do States Agree to Arms Transfer Controls?

The question why states agree to arms transfer controls has so far not received much attention from scholars. The application of several international relations theories to arms transfer control has not been fruitful. The application to arms transfers has been done for liberal institutionalism, collective security and critical theory (Mearsheimer 1994; Bromley/Broszka 2008; Gärtner 2010; Erickson 2015a). Central to why these theories do not fit nicely to arms transfers is that, in many cases, there is a gap between commitment and compliance. Regarding the EU CP, there have been attempts to apply the notion of normative power Europe (NPE) developed by Manners in 2002.

(9)

5 The concept of NPE includes that first, the EU is constructed on a normative basis and second, that this normative basis predisposes it to act in a normative way in world politics (Manners 2002: 252). Hansen and Marsh (2014: 266) argue, that the values on which the NPE assumptions are grounded can also be found in the EU arms control regime. The five core values can for example all be found in some way in the consolidated criteria of the CP (ibid.). Also in line with the notion of NPE is the importance of harmonisation efforts accompanying the CP such a User`s Guide, a list defining military equipment or the requirement of bilateral consultation when a state considers to grant a licence that is `essentially identical` to a licence that has been denied by another state in the past three years (Hansen/Marsh 2014: 269). Taken together, states would be expected to comply with the arms control norms and thus apply the CP consistently. Hansen and Marsh (2014: 267) show with their study on EU arms exports to Libya, that NPE is an exaggerated construct which does not hold true when ideals meet interests. This finding is supported by a study from Erickson (2011). She examines the relationship of EU arms exports and human rights, conflict and democracy and finds that countries with poor human rights records often received arms from EU member states. The gap between commitment and compliance has led Erickson to focus on reputation in explaining arms transfer control in her 2015 book “Dangerous Trade: Arms Exports, Human Rights, and International Reputation”. The notion of reputation seems useful because it acknowledges the gap between commitment and compliance which is the main obstacle for NPE and other international relations theories in explaining arms transfer controls.

The starting point of Erickson`s theoretical discussion of arms transfer controls is the 2013 ATT which sets for the first time worldwide, legally binding humanitarian arms trade standards (Erickson 2015a). These `responsible` arms transfer standards bring costs without benefits which makes it puzzling why states agree to them. Erickson argues that states strategically adopt `popular policies out of social concern for their international reputation`. While in other international relations theories reputation is seen merely as a means, for example to make credible threats, she conceptualises reputation also as an end in itself. By conceptualising reputation not only as a means but also as an end, Erickson acknowledges the social dimension to it. Norm adoption is done to avoid social costs and gain the social benefits of a good reputation such as international legitimacy (Erickson 2015a: 18).

At the international level, reputation can work as a social incentive and motivate leaders to commit to popular initiatives although material costs are present (Erickson 2015a: 24). States, as rational actors in a social context, care about their reputation first, because it positively reinforces their identity and second, because of the implications for their social standing within the international system (ibid.: 25). Not only small or weak states can be interested in a good social standing to make up for little military or market power, but also more powerful embedded states can. Erickson writes that the concern of states for their reputation is a “search for recognition of their equal status as `good` members of the international community and an effort to increase their standing in that community” (Erickson 2015a: 27). On a less abstract level, reputation is formed in the context of international institutions and carried by diplomats and elites meeting as representatives of their countries. Reputation can translate into social influence and

(10)

6

moral authority (Erickson 2015a: 29). This makes reputation further valuable, also for powerful states, because it can be a cheaper and subtler means to influence other states than with hard power.

While reputation can work as an incentive for states to adopt popular policies, there is a key limitation to reputation when it comes to the behaviour of a state (Erickson 2015a: 30). Regarding arms transfer controls, commitment to policies is observable to other actors. In contrast, practice is far less observable which limits the ability of reputation to influence behaviour. With missing accountability mechanisms, adoption of popular policies can bring the reputational reward without the costs of implementation. This helps to explain the commitment-compliance gap with which other theories often struggle. In international politics, such a gap does not necessarily lead to penalties because there often is little transparency and a lack of enforcement capabilities (ibid.: 18). Erickson argues that while commitment may serve an international audience, compliance is relevant for the domestic audience (ibid.: 19). In the domestic realm, the interest in arms transfers is small which makes it uninteresting for politicians to advocate strict arms export controls (Erickson 2015a: 31). If politicians do promote strict arms controls, they are confronted with arguments on jobs and economic and security benefits. This makes it more rewarding to promote commitment to arms controls rather than compliance, also in the domestic realm. Erickson argues that it is scandals which work with reputation to generate compliance. This can be through scandal outbreaks, which forces politicians to improve practice and repair reputational damage, often with highly visible action. An increased threat of scandals can also push decision makers towards compliance (ibid.: 23). A scandal does not need to result from illegal action and states reactions to scandals can be diverse. For this function of scandals to work, however, it is necessary to have transparency and a public who is willing to make use of the information. This task has been taken up by NGOs with improvements in arms trade transparency in the 1990s. By making specific cases of irresponsible arms exports public, NGOs can point out the gap between the self-image of a government as a responsible actor and the irresponsible practices (ibid.: 35). With `naming and shaming`, NGOs can exploit the governments` interest in avoiding bad press at home and can thus be seen as a “direct conduit between reputation and practice in domestic politics” (Erickson 2015a: 37).

This function of `naming and shaming` can be expected to work primarily in democracies, where transparency and accountability are a valued part of governance (Erickson 2015a: 19). Although arms trade scandals do not decide elections, they can nevertheless strengthen the opposition or weaken the legitimacy of the government and trustworthiness of leaders (ibid.: 33). These consequences, Erickson argues, come from reputational damage for the actors deemed responsible for the scandal. It can therefore also be that states adopt norms because of domestic pressure and not in order to improve or keep up the reputation in international politics. These two possibilities are, however, linked to each other. Not only can reputational damage, resulting from scandals, push states towards the adoption of `responsible` arms transfer controls but it can also make a state become a leader in promoting stricter controls (Erickson 2015b). This can happen if a state is forced to adopt new norms through domestic pressure. Because these new norms bring along costs which other states do not face, there is an incentive to diffuse these norms in order to `level the playing field` (ibid.).

(11)

7 Although the notion of reputation is helpful in explaining the commitment-compliance gap, there are some weaknesses to it. First of all, the assumption that stricter arms controls are only supported because of reputational gains or domestic pressure is rather pessimistic. Can a state not be in favour of stricter arms controls without ulterior motive or domestic pressure? This consideration points to another weakness which is that the state is conceptualised very homogenous. Arms transfer controls are, however, very complex and involve often different administrative branches up to the political leaders of a country in cases of big or controversial deals. Especially in democracies, on which Erickson focuses, this diversity is bound to result in different views on how to implement a treaty and whether or not to support stricter arms controls. A third weakness is that Erickson leaves aside the `classical` explanations for why states export arms. This is unfortunate as these factors are likely to influence the success of the domestic realm to narrow the commitment-compliance gap.

Looking at the CP and EU arms exports to Saudi Arabia, the theoretical notion of reputation has some valuable implications. Generally, the implementation of the CP can be expected to be inconsistent because the implementation is left to the individual states in which domestic circumstances would influence the compliance. States in which arms trade scandals are scarce can be expected not to comply with the CP while states prone to scandals can be expected to implement the CP more strictly. The same pattern should be found regarding NGOs. Countries with few NGOs dealing with the arms trade are expected to comply with the CP less than countries with many or dominant NGOs. Looking at the costs of implementing the CP, the implications are not that clear. A state of which Saudi Arabia is an important arms export destination can be either expected to comply to the CP more but also less strictly. The argument speaking for a less strict implementation would be that any denial of a licence application could threaten the relationship. But speaking for a stricter implementation is that a country which sells many weapons to another country has also more leeway to deny one or two licences without having to relinquish the benefits of arms exports. A denial of one out of many licences could, however, also be seen as a sign of whitewashing the export control mechanisms because many other licences were granted which could make the loss insignificant. These implications are revisited during the discussion of the research questions in chapter seven.

3 Arms Control

During the Cold War, attempts for international conventional arms control were largely unsuccessful and the UN Charta, with article 51 laying out states` right for self-defence, “provided an almost blanket excuse for arms transfers” (Wezeman 2008: 195). Concerning weapons of mass destruction (WMD), however, multilateral treaties were achieved. In 1970 the `Treaty on the Non-Proliferation of Nuclear Weapons` entered in to force and was followed five years later by the `Biological and Toxin Weapons Convention`. After the Cold War, in 1997, the `Chemical Weapons Convention` entered into force (Bauer 2010: 306). Regarding inhumane weapons, there is the `Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons which may be Deemed to be Excessively Injurious or to have Indiscriminate

(12)

8

Effects` (CCW), which entered into force in 1983. The CCW now encompasses five protocols and was amended in 2004 to also cover internal conflicts. Saudi Arabia is party to the CCW but did not adopt Protocol II, which covers anti-personnel landmines and booby-traps, and also did not adopt the amendment which means that internal conflicts are not covered by the CCW (UNOG: CCW).

Towards the end of the Cold War, negotiations started on the `Treaty on Conventional Armed Forces in Europe` (CFE) which sets limits on certain categories of conventional military equipment and includes provisions to destroy excess weaponry aimed at the stockpiles which were built up during the Cold War (Bauer 2010: 306). The CFE came into force in 1992 but is now in a major crisis (ibid). Russia suspended the implementation of the CFE in 2007 and later suspended the participation in 2015 (Arms Control Association 2015). Around the same time of the start of the CFE, the UN Register of Conventional Arms (UNROCA) was introduced. Since 1991, governments can voluntarily provide information about imports and exports of major conventional arms (since 2006 also small arms and light weapons [SALW]) as well as military holdings, domestic procurement and policies (Bauer 2010: 309). The UNROCA is aimed at building confidence and cooperation between states and promoting transparency norms (Holtom et al. 2011). The highest number of states submitting reports to UNROCA was 126 in 2001. The number declined since then to 47 states reporting to UNROCA in 2015 for the calendar year 2014, however, some reports might still follow (UNODA: UNROCA). Since states report imports and exports from and to other states, information can also be found for states not submitting reports. This is for example the case for Saudi Arabia, which has never submitted a report.

Regarding arms transfer controls, the end of the cold war brought about changes which Bauer categorises into four shifts or trends. During the cold war period, the main function of arms transfer control was to deny transfers to the opposite bloc (Bauer 2010: 312). Arms exports of the Western Bloc were regulated by the Coordinating Committee on Multilateral Export Controls. In the 1990s, this ideologically grounded transfer control system transformed to a system where agreed guidelines are being assessed on a case-by-case basis. While loopholes and scandals pushed governments towards more restrictive export control policies, economic considerations gained importance for these same policies. This shift is also reflected in the transformation of the Coordinating Committee on Multilateral Export Controls to the Wassenaar Arrangement in 1995, which includes many former Warsaw Pact countries, including Russia. The non-legally binding Wassenaar Arrangement established guidelines for decision-making about the export of conventional weapons and dual-use goods and technologies (Wassenaar Arrangement: About us). The second shift is that the territorial focus of former export policies was replaced by a focus on end-user and non-state actors as a response to risks such as re-exports, diversion or terrorism (Bauer 2010: 312). A third trend is the increasing complexity of export controls on the one hand of items (goods and services) and on the other hand of activities such as export, trans-shipment or brokering. The fourth trend identified by Bauer is that, while some controls were getting stronger, controls in trusted communities were loosened which resulted in more tailored control mechanisms, allowing economically beneficial arms production (Bauer 2010: 313).

(13)

9 Besides these four shifts and a general development towards conventional arms transparency, in the 1990s, there was also a normative shift leading to the emergence of `responsible` arms transfers (Erickson 2015a: 61). With `responsible` arms transfer policies there was a link established between arms transfers and the broader economic, social and human security problematic taking into account human rights and the internal policies and practices of the recipient country. Erickson argues that this shift was partly related to the emergence of human rights and humanitarian issues in the global spotlight after the Cold War and more specifically to four developments linked to the arms trade in the 1990s (ibid.).

The first development was the 1991 Gulf War in which coalition soldiers were confronted by large numbers of weapons their governments sold to Iraq during the 1980s. To prevent such an unnoticed build-up, the UNROCA was established as an early-warning mechanism. The Gulf War also revealed the poor control mechanisms of arms exporting states and led to several domestic scandals.

A second development were civil and ethnic conflicts after the Cold-War which demonstrated the detrimental effect of SALW (Erickson 2015a: 62). For this reason, the Group of Governmental Experts on Small Arms was introduced by the UN in 1995. In 2001, the `UN Programme of Action to Prevent, Combat and Eradicate the Illicit Trade in Small Arms and Light Weapons in All its Aspects` (PoA) was established. Focusing on the illicit trade helped to generate support but only addresses the problem partly. This is because most of the SALW in the illicit trade were once produced under government control (Marsh 2002), which makes it important to improve transparency and control of the legal trade (Bauer 2010: 310). Bauer sees the main achievement of the PoA in putting the issue on the agenda and raising its profile.

A third development which led to the emergence of `responsible` arms transfer controls was the `International Campaign to Ban Landmines` and the 1997 `Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on their Destruction`, also known as Ottawa Treaty. The key to the success of the Ottawa Treaty is simultaneously also its biggest weakness, namely that it was taken out of the UN framework to circumvent the objection of member states, but also mine producing countries, such as Russia and the USA (Bauer 2010: 308). Erickson attributes to the landmine campaign the “marriage between traditional security and human security” which has led to a “growing consensus on `good` conventional arms policy linked to humanitarian values” (Erickson 2015a: 63).

The fourth development Erickson identifies is the growing role of NGOs linked to arms transfer controls. NGOs pushed arms transfer controls on the political agenda and, through the improvements in transparency, were able to hold states accountable. Erickson concludes that “within a single decade, the normative environment for the global arms trade has changed noticeably, from non-interference on behalf of national material interests to `responsibility` and humanitarianism” (Erickson 2015a:63).

The shift to responsible arms transfer controls, resulting from the four developments, led amongst others to the creation of the EU CoC in 1998 which is discussed in the next section but also to the non-legally binding Principles Governing Conventional Arms Transfers through the Organisation for Security and Co-operation in Europe (OSCE 1996). The legally binding ATT which entered into force in 2014 is to date the

(14)

10

most far reaching outcome of this shift. The ATT includes small arms and light weapons as well as provisions on human rights and international humanitarian law. It has many similarities to the EU Common Position, not least because the EU and its member states took an active role in the creation of the ATT. Saudi Arabia has not signed the ATT (UNODA: Arms Trade Treaty).

3.1 Common Position

The emergence of `responsible` arms transfer controls in the EU is reflected in the adoption of the CoC on Arms Exports in 1998 (EU Council 1998). The CoC set eight criteria, based on agreed criteria by the EU Council in 1991/1992, which member states should apply before granting an export licence. In the CoC it was stated that the purpose is “to set high common standards which should be regarded as the minimum for the management of, and restraint in, conventional arms transfers by all Member States, and to strengthen the exchange of relevant information with a view to achieving greater transparency” (EU Council 1998). Although the CoC was aimed at harmonizing EU member states arms export controls, the final control over licencing decision remained in the hands of the states.

In 2008, the CoC became a legally binding Common Position (CP) with the title `Council Common Position 2008/944/CFSP of 8 December 2008 defining common rules governing control of exports of military technology and equipment` (EU Council 2008). With the introduction of the CP came some changes to the consolidated criteria guiding export decisions as well as an obligation for states to produce national reports (Bromley 2012: 5). The final control over arms export decisions is still maintained by the EU member states. The eight criteria laid down in the CP are as follows1_:

1. “respect for Member States’ international obligations and commitments, in particular the sanctions adopted by the UN Security Council or the European Union, and agreements on non-proliferation and other subjects;

2. respect for human rights in the country of final destination as well as respect by that country of international humanitarian law;

3. the internal situation in the country of final destination – Member States will not allow exports that would provoke or prolong armed conflicts or aggravate existing tensions or conflicts in the country of final destination;

4. preservation of regional peace, security and stability;

5. security of Member States and of territories whose external relations are the responsibility of a Member State, as well as that of friendly and allied countries;

(15)

11 6. behaviour of the buyer country with regard to the international community, as regards in

particular its attitude to terrorism, the nature of its alliances and respect for international law; 7. existence of a risk that the military technology or equipment will be diverted within the buyer

country or re-exported under undesirable conditions;

8. compatibility of the exports with the technical and economic capacity of the recipient country, taking into account the desirability that states should meet their legitimate security and defence needs with minimal diversion of human and economic resources for armaments” (EU Council 2008).

Licence applications are assessed on a case-by-case basis against these eight criteria. Member states should deny a licence if the assessment of the first four criteria is not favourable while the criteria five to eight ònly` have to be taken into account in considering a licence application (Bromley 2012: 3). States are obliged to inform other member states about denied licence applications with an explanation why it has been denied (CP, Article 4). Before another state wants to grant an èssentially identical` licence which has been denied in the past three years, it needs to consult the state or states which denied the licence and inform it/them about the final decision. What èssentially identical` means is left to the interpretation of the states. Publicly available is ònly` the `Common Military List of the European Union` (ML) containing 22 categories of military equipment which are used to classify licence applications (EU Council 2015). The 22 categories2_{are, however, still broad enough that a licence concerning one category can be granted}

while another licence can be denied without breaching the CP.

There is a Working Party on Conventional Arms Export (COARM), consisting of governments officials, which meets regularly (Bauer 2010: 315). The aim of COARM is to harmonize the application of the CP. An important document produced by COARM is the `User’s Guide to Council Common Position 2008/944/CFSP Defining Common Rules Governing the Control of Exports of Military Technology and Equipment` (EU Council 2015: User`s Guide). The CP User`s Guide is intended to help states apply the CP by providing agreed guidance for the interpretation of the criteria and articles. Since 1998, every year a report about the implementation of the CP is published by the EU Council, which is, according to Bauer, one of the main tools to assess how states interpret and apply the CP (Bauer 2010: 315). These yearly reports have become more detailed over the years while in the beginning they only contained aggregated data for each member state. Since 2004, the reports include for each member state, inter alia, the number and worth of licences issued per ML category and destination country. Licence denials are reported per destination country and ML category but only aggregated for all member states. The annual EU council reports were not intended to be publicly available when the CoC was introduced in 1998. In 1999 it was agreed, however, that the reports are made public after the insistence of the Finnish EU presidency and pressure from the European Parliament and NGOs (Bauer/Bromley 2004: 5).

(16)

12

Besides the EU member states, there are several other countries3_{which have officially aligned themselves}

with the consolidated criteria and principles of the CP. These countries are, however, not included in the annual EU Council reports. In November 2015 the EU parliament adopted a resolution on the implementation of the CP (EU Parliament 2015). The resolution states that the CP is being applied “loosely and interpreted inconsistently by the Member States” and there should arrangements be made to make arrangements for independent checks and penalties if the CP is infringed (ibid.: 9). The parliament further notes that there is a need to further clarify Criterion 2. An introduction of a right to challenge granted licences by other member states is proposed as well as that the format of the EU Council reports is changed to a searchable online database (ibid.: 9; 12).

4 Method and Data

To analyse the application of the CP, a case study on Saudi Arabia is conducted. A qualitative approach with a focus on one destination country seems appropriate because of two reasons. First, the application of the CP is done on a case-by-case basis which means that each of the 28 EU member states assesses, or is supposed to assess, every single licence against the consolidated criteria. Together with the 22 ML categories, this leaves substantial room for variability which makes it necessary to also include basic information on the side of the exporting states and not just on the side of the importing state. This reason alone does not speak against a quantitative approach. The second reason, however, is that denials of licences are scarce (only less than 1% of an average of over 4000 licence applications per year is reported to be denied) and a denial does not necessarily correspond with a restrictive export policy and vice versa. This combination of room for variability on the one side and only few denials on the other, seems to make a qualitative case study a suitable approach.

The main data source are the annual reports on the implementation of the CP published by the EU Council (the reports can be found on the EU External Action Website). There are so far 17 reports covering the years 1998 to 2014. Because these reports only contain aggregated data on licence denials, national reports are also included. National reports which are more detailed than the annual EU Council reports were identified with the help of the “SIPRI Fact Sheet: National Reports on Arms Exports” from 2011 which contains information on which national reports contain what kind of information (Weber/Bromley 2011). All the reports included here can be retrieved from the SIPRI National Reports Database. Only national reports which report at least licence denials per destination country were included in order to be able to deduce whether there were any licence denials regarding Saudi Arabia. Reports published in another language than English, German, French or Dutch were searched keywords and the relevant section subsequently translated with Google Translator (www.translate.google.com). These reports were only included when there was no doubt about the content. Because reporting of licence data is often done in

3_{The countries which aligned themselves with the CP are: Albania, Bosnia and Herzegovina, Canada, the former} Yugoslav Republic of Macedonia, Georgia, Iceland, Montenegro and Norway.

(17)

13

Data source: SIPRI National Reports Database. This list does not aspire to be comprehensive.

the form of tables, translating was rather unproblematic. An overview of national reports which contain information on licence denial per country can be found in Table 1.

Table 1. National Reports with Information on Licence Denials per Destination Country

Country Years Covered Country Years Covered

Belgium Romania 2000-2014 Flanders 2004-2014 Slovakia 2007-2014 Wallonia 2003-2014 Spain 2004-2014 Denmark 1999-2011 2013-2015 Sweden 2001-2015

Germany 1999-2014 United Kingdom 2008-2015

Netherlands 1998-2014

With the EU Council reports and the national reports, it is possible to establish on the one hand what licences have been granted, broken down by ML category, and on the other hand what licences have been denied with the cited criteria of the CP on which the denials are based upon. Especially the data on denials is valuable because it shows which ML categories were subject to objection with a cited reason in the form of CP criteria. To better understand the reason lying behind the denial, the CP User`s Guide will be drawn upon with a focus on the most cited criteria which led to a denial. Considering the theory, however, other factors are expected to be decisive in explaining why a state complies with the CP. These factors mainly entail past arms transfer scandals or the threat of such scandals, which is linked to the presence of NGOs committed to the issue. Data sources allowing to explore these factors comprise of scientific literature, newspaper articles and NGO reports.

Another important factor is the trade relationship, or the value of traded arms, between individual EU member countries and Saudi Arabia. This can be established on the basis of licence applications reported in the EU council reports. It is, however, more comprehensive to also include other arms trade data in the form of actual exports because of two reasons. First, states are supposed to make full submissions of arms trade data to the EU council reports, which would include the value of actual exports. Because this is not done consistently, important information is missing for example for the UK. Second, the EU Council Report started in 1998 but a longer perspective can also be illuminating. This second arms export data source is the SIPRI Arms Transfer Database (ATD). The ATD contains information on actual deliveries only of major conventional weapons because public information is inadequate for tracking all arms transfers (SIPRI: Methodology). To make the data comparable, SIPRI has developed the trend-indicator value (TIV) which is based on known production costs of a core set of weapons and thus corresponds more to the transfer of military resources rather than the financial value of the transfer. Data on the military expenditure of Saudi Arabia is retrieved from the SIPRI Military Expenditure Database (MED).

(18)

14

The export data is relevant to establish the export relationship between EU member states and Saudi Arabia but it is the reported data in the EU Council reports which make an assessment of the CP possible. This assessment is based on licencing data to which there are some important limitations which need to be mentioned here. A first limitation is that the value of a granted licence does not need to correspond to the value of the actually exported goods. This means that looking at the value of licences can mean overestimating the importance of the deal. A second limitation is that it can be that licence applications which have no chance of being granted do not make it to the official licencing procedure and are thus not reported (Bauer/Bromley 2004: 16). In the case of the UK, the export regulations exempt government-to-government deals from the requirement of a licence and are thus also not reported (ibid.: 11). The UK also issues open-licences which cover multiple shipments but do not specify the value (ibid.) which hampers the interpretation of the reported licencing data. For Germany, data on licences covering multiple shipments is not reported to the annual reports (ibid.). Together, these limitations underline the need for a cautious interpretation of licence data and especially denials of licences. Nonetheless are licencing data useful to assess the implementation of the CP, not least because they reflect recent policy choices, but the addition of actual export data enables a fuller understanding (Bauer/Bromley 2004: 12).

The proceeding part of the paper is structured as follows. In chapter five, an overview of military expenditure data of Saudi Arabia is given, combined with military imports of Saudi Arabia and exports from the EU. This helps to establish trade relationships and procurement trends of Saudi Arabia based on data provided by SIPRI. Chapter six, central to this case study, addresses the implementation of the CP and thus the research questions. This part grounds mainly on data of the EU Council reports and data from the relevant national reports but NGO reports and Newspaper articles will be drawn on too. Looking at denied licences, the focus lies on possible arms export scandals which stems from the theoretical notion of reputation. The third part in chapter seven consists of a discussion of the research questions in light of the results from the case study.

5 The Arms Trade Between EU Member States and Saudi Arabia

Saudi Arabia has become the biggest arms importer in 2015 with imports worth $3.2 billion (Source: ATD). Also in terms of military expenditure, displayed in Figure 1, have there been huge increases since the early 2000s. Although far behind the USA ($596 billion) and China ($215 billion), Saudi Arabia ranked third in 2015 with a military expenditure of over $87 billion. Measured as a percentage of the Gross Domestic Product (GDP), Saudi Arabia spent 13.9% on military expenditure in 2015 and an average of 8.9% for the years 2005-2015. Only Oman is reported to spend a greater share of the GDP on the military with an estimated 16.2% in 2015 (Source: MED). A high military expenditure rate measured as a percentage of the GDP can be found for almost all Gulf states with figures much higher than the global average of around 2.5% (Solmirano/Wezeman 2010).

(19)

15 Looking at the patterns of Saudi military spending, Attar (2010) identifies four phases. The first phase is the Egyptian-Saudi reconciliation, beginning in 1968 and halting in 1977 due to improvements in the connection between Egypt and Israel. Increases in military expenditure during that phase could be explained by financial assistance to Egypt (Attar 2010: 114). The second phase is marked by the Soviet invasion of Afghanistan lasting from 1978 to 1989 and the Iranian Revolution in 1979. These two events led to a sharp increase in military purchases. The 1991 Gulf War, accompanied by the presence of US forces in Saudi Arabia, is the third phase which also contributed to military imports. The fourth phase is marked by the terrorist Attack of the 11 September 2001 and the subsequently initiated `war on terror` (Attar 2010: 2015). Since then, Saudi Arabia has recorded huge increases in military expenditure parallel to increased instability in the region. Developments contributing to this instability include internal and external terrorist threats, the Arab Spring with the intervention of Saudi Arabia and the United Arab Emirates in Bahrain in March 2011 (Nuruzzaman 2013), the Saudi-Arabia led military intervention in Yemen and tensions between Saudi Arabia and Iran (Cordesman/Burke 2016).

Arms imports make up only a fraction of the military expenditure but nonetheless added up to TIV 3.1 billion in 2015. This number is not only the highest so far for Saudi Arabia but the highest for any country in 2015. The development of Saudi-Arabian arms imports can be found in Figure 2. The increase in arms imports in the 1970s and high numbers of imports until the end of the 1990s as well as the increase in imports starting in the mid-2000s can be attributed to the four phases of military expenditure and the regional instabilities described above. Imports after the Gulf War can further be linked to orders made in reaction to the Gulf War and perceived threats from Iran and Iraq (Cordesman/Obaid 2004). The numerous arms imports strained the budget and even led to a budget crisis in the mid-1990s. Saudi Arabia was able to pay some ordered arms with oil but declining oil revenues and the `oil-crash` in late 1997 forced Saudi

0 10 20 30 40 50 60 70 80 90 Con sta n t 2014 Bil lion US$

Data Source: SIPRI Military Expenditure Database. Figures represent the adopted budget and not the actual expenditure which is likely to be higher. Gaps in the line are a result of missing data.

(20)

16

Arabia to cut down in arms purchases (ibid.: 20). This cutback can be seen especially for the years 2000 and 2001 with the lowest imports for around 30 years.

Saudi Arabia was long completely dependent on arms imports and is now only making slow progress in developing its own arms industry (Cordesman/Obaid 2004: 18). Saudi Arabia has made progress in the maintenance of arms and can build small arms and ammunition which consist, however, mainly in assembling imported parts. There have also been `offset` deals, some of which resulted in a reduction of technology imports but many of them rather were a symbolic effort to employ Saudis (ibid.). Saudi Arabia imported most of the arms from the USA which make up 53% of the total Saudi arms imports between 1998-2015. The UK and France are also major suppliers of arms to Saudi Arabia. Between 1998 and 2015, Saudi Arabia imported 24% of its arm from the UK and 10% from France which makes the two countries the second and third biggest suppliers. Other EU member states in the top ten list of arms suppliers for Saudi Arabia are Spain on rank four accounting for 3%, Germany on rank six accounting for 1.59%, Sweden on rank 9 accounting for 0.86% and Belgium on rank ten accounting for 0.81%. Arms imports by Saudi Arabia from all EU member states made up 42% between 1998 and 2015.

The picture looks somewhat different when looking at these countries` share of arms exports going to Saudi Arabia. Between 1998 and 2015, US arms exports to Saudi Arabia accounted for 6.8% of the total which makes Saudi Arabia the second biggest recipient of US arms after South Korea (9.6%). An overview of EU member states` arms recipients of which arms exports to Saudi Arabia make up a substantial part of the total can be found in Table 2. For Germany and Sweden, which both are in the top ten list of Saudi arms imports, Saudi Arabia is only a minor export recipient. Between 1998 and 2015 Saudi Arabia was on rank 24 of German arms exports accounting for 0.95% of the total arms exports and on rank 16 of Swedish arms exports making up 2.1% of the total.

0 0.5 1 1.5 2 2.5 3 3.5 1952 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 TIV ( Bil lio n s)

Figure 2. Saudi Arabian Arms Imports

Data Source: SIPRI Arms Transfer Database. Unit of measure is the trend-indicator value (TIV). Only major weapons are included.

(21)

17

Table 2. Top Recipients of EU Member States` Arms Exports 1998-2015

Supplier Recipient (share of suppliers` total exports)

United Kingdom 1. Saudi Arabia (21.8%), 2. USA (14.6%), 3. India (7.7%)

Belgium 1. Jordan (30.2%), 2. Bulgaria (27.2%), 3. Saudi Arabia (13.6%) Spain 1. Norway (24,2%), 2. Australia 16.4%), 3. Saudi Arabia (6.7%)

France 1. United Arab Emirates (16.9%), China (10.5%), 3. Saudi Arabia (6.2%)

The UK and the USA are the countries with the closest ties to Saudi Arabia. Both countries on the one side account for a substantial part of Saudi arms imports, and on the other side sell a large share of total arms exports to Saudi Arabia. The UK is, however, the country of which arms exports to Saudi Arabia make up the largest share of the total exports adding up to over 20%. France and Spain are in a similar position but exports from France have more weight for Saudi Arabia and account for 10% of the total Saudi imports compared to 3% in the case of Spain. This difference is the result of more overall arms exports of France ranking fifth on the list of the largest arms exporter worldwide through the years 2009 to 2013 (Wezeman/Wezeman 2014). The largest share during the same period is made up by the USA with 29%, followed by Russia with 27%, Germany with 7% and China with 6%. After France on rank five with a share of 5%, is the UK with 4% and Spain with 3%. Belgium is a special case in the list as the 0.81% share in Saudi arms imports translate into 13.6% of the total Belgium arms exports. This shows that it is important to take into account both, the share of exports going to a certain recipient, and the effect these exports have on the composition of the arms imports of the recipient state. Still somewhat different is the composition of the EU member states which granted the most licences under the CP. This is discussed in the next chapter on the implementation of the CP.

6 Implementation of the Common Position

The most recent EU council report included here is the seventeenth report covering the year 2014 which was published in May 2016. The first few reports were published in December of the year following the reported data which meant that licences issued early in a year, were reported almost two years later. The reports covering the years 2005 and 2006 were publish a bit earlier, in October of the following year. Since then, however, the time lag between reported data and the publishing date of the reports has grown to 17 months in the case of licences issued in December of the reported year or 29 months in the case of licences issued in January of the reported year. This means that it can take up to almost two and a half years for a licence to be reported. Not only does this hamper the possibility to track the governments` decisions but many of the arms for which licences have been granted may already be shipped when the report is published. Most countries publish their national reports, however, less lagged.

Data Source: SIPRI Arms Transfer Database. Only countries of which arms exports to Saudi Arabia accounts for more than 5% of the total arms exports 1998-2015. Only major weapons are included.

(22)

18

Not only are the EU Council reports published with an increasing time lag but many countries fail to make full submissions to the report. This mainly concerns data on actual exports which many countries (9 out of 28 for the seventeenth report) do not report, notably including the three biggest arms exporting EU countries France, Germany and the UK (Bromley 2010: 22). There have been no real improvements in the last year which is why aggregated data on actual exports is since the fourteenth report not provided anymore. An exception are destination countries for which all exporting countries submitted data. A reason for states not to submit actual export data is that this data is not collected at the national level which could be done by requiring states to submit reports on how and when issued licences were used (Bauer/Bromley 2004). Not all states have such a requirement and are unwilling or unable to introduce it (Bromley 2010: 23).

6.1 Issued Licences

It is possible to establish the number and value of licences issued to Saudi Arabia since 2001 and up to 2014. In this period 8978 licences were granted for exports to Saudi Arabia with a total value of €34.7 billion. This results in an average of 641 granted licences worth €2.5 billion per year. In 2009 most licences were granted with a value of roughly €5 billion. In 2010 granted licences were close to the average with a value of €2.54 billion but the value of granted licences increased again for the years 2011-2014 with values around €4 billion. France has by far issued the most licences, accounting for 51% of the value of all granted licences or €17.2 billion. With issued licences worth €7 billion the UK comes second to France and accounts for 20% of all issued licences. It needs to be noted that in the UK government-to-government deals do not require a licence which probably best explains the discrepancy between data of granted licences and actual exports for which the UK shows much higher numbers than France. The biggest government-to-government are the al-Yamamah deals for discussed later on. Belgium, Germany and Italy each issued licences worth around €2.7 billion accounting for a share of 7%.

Since 2005, granted licences are not only reported broken down by destination country but also by ML category. The total value of licences issued per ML category for the years 2003-2014 can be found in Figure 3.The biggest share of almost a third of the total (€9.3 billion) is made up by ML10 which covers aircrafts, unmanned airborne vehicles as well as related parts and equipment. On rank two and three are ML9 and ML4, both of which make up around 12% of the total (€4 billion). Category ML9 comprises vessels of war and special naval equipment as well as components thereof. Category ML4 includes bombs, torpedoes, rockets, missiles and other explosive devices as well as related equipment. The majority of licenses for ML4 were issued by France with a share of 80%.

(23)

19 On rank four of the ML categories of which the highest values were issued is ML3 which includes ammunition and fuze setting devices, which have the function of detonating the explosive material of a munition, for example for artillery. Over a third of the licences for ML3 were issued by France in 2014 with a value of over €1 billion and looking at the total between 2003 and 2014 France issued 63% of all ML3 licences. Another big share of ammunition supply to Saudi Arabia is made up by licences issued by Belgium which add up to 20% of all issued ML3 licences. Category ML6 covers ground vehicles and components. Also regarding ML6, France issued most licences accounting for 53% of all ML6 licences. Austria issued two ML6 licences worth €433 million in 2009 which make up 18% of all ML6 licences issued. On rank six are licences covering ML5 which comprises fire control, related alerting and warning systems as well as test, alignment and countermeasure equipment. Regarding ML5, most licences were issued by Germany, accounting for 46% of the total (42% alone were issued in 2012) and France which issued 44% of all ML5 licences between 2003 and 2014.

Regarding ML10 (Aircraft) and ML9 (Vessel) there are two major deals between Saudi Arabia and the UK and France respectively. In 1988 the UK signed the al-Yamamah deal with Saudi Arabia worth around $7 billion (Source: ATD). The deal included 72 Tornado fighter aircraft, 30 Hawk training aircraft as well as other training aircrafts and missiles. The deal was renewed in 1993 for another 48 Tornado aircraft termed al-Yamamah-2. The al-Yamamah deal was lobbied for by Margaret Thatcher who guaranteed full discretion to the Saudi Royal Family (The Guardian: The BAE Files). The deal was made between the two governments with the British defence company BAE as the main contractor. The deal got associated with corruption allegations that illicit payments were made by BAE to Saudi royals in order to secure the contract (ibid.). These allegations were triggered by investigations by journalists of The Guardian. An investigation by the

€ -€ 1.00 € 2.00 € 3.00 € 4.00 € 5.00 € 6.00 € 7.00 € 8.00 € 9.00 € 10.00 ML 1 ML 2 ML 3 ML 4 ML 5 M L6 ML 7 ML 8 ML 9 ML 10 ML 11 ML 12 ML 13 ML 14 ML 15 ML 16 ML 17 ML 18 ML 19 ML 21 ML 22 Bi lli o n E u ro s

EU Military List Category

Figure 3. Issued Licences to Saudi Arabia 2003-2014

Data source: EU Council annual reports. No licences were issued for category ML20. Values of licences for the ML categories where no columns are visible vary between €6 million and €63 million.

(24)

20

Serious Fraud Office was halted in 2006, supposedly resulting from pressure of the Saudi government, notably during a period when a new order of aircrafts was on the table. The deal was successfully renewed for a third time in 2006 but under a new name: Project Salam (Source: ATD). Project Salam includes 72 Eurofighter Typhoon4_{which were delivered between 2009 and 2015. The three deals add up to an}

estimated £40 billion and mark the biggest arms deal in the UK history (The Guardian: The BEA Files). To settle US and UK investigations into the corruption allegations of the al-Yamamah deals as well as other corruption allegations, BAE agreed to pay a penalty of almost £300 million (ibid.).

France made a similar deal with Saudi Arabia but not for the Royal Saudi Airforce but for the Royal Saudi Navy. The so called Sawari deal worth $2.5 billion included four frigate ships and two support ships which were delivered in the mid-1980s as well as 20 helicopters and missiles for the ships (Source: ATD). As with the al-Yamamah deal, the Sawari deal was renewed in the 1990s as Sawari II worth $3.4 billion mainly including three frigate ships delivered between 2002 and 2004. In June 2015, a civil and military agreement was reached between France and Saudi Arabia encompassing a deal worth around $12 billion (Al Arabiya 2015). A few months later it was announced that the construction of a naval research centre would be discussed which would include a working group to study future naval projects, possibly including a third Sawari deal (Defense News 2015a). Saudi Arabia has further funded Egypt in buying French arms. With these funds Egypt, was able to buy from France amongst others two helicopter carriers, four corvettes and a frigate ship (Defense News 2015b).

A third big deal Saudi Arabia made is the so called Miksa (Ministry of Interior Kingdom of Saudi Arabia) deal agreed in 2009 (Defence Talk 2009). The deal was signed with EADS Defence and Security, a subsidiary of the French-German conglomerate EADS which since 2013 is called Airbus Group. The Miksa project goes back to the early 1990s because Saudi Arabia was thinking about securing the boarder to Iraq after the Gulf War (Defence Talk 2009). The aim of the Miksa deal is to enhance boarder security with security posts, monitoring of Saudi land and borders as well as a roughly 9`000 km long security fence on the border (ibid.). The deal includes interceptor vessels and patrol ships from France and Germany (Mer et Marine 2015; ATD). The value of the deal is not known but has been estimated to be around €2 billion (Le Point 2009). Germany issued licences for border security (mainly ML5 licences) in 2012 worth over €1.1 billion which accounts for a share of 41% of all granted licences to Saudi Arabia between 2001 and 2014 and made Saudi Arabia the biggest recipient of granted licences by Germany in 2012 (Rüstungsexportbericht 2012).

It is not surprising that the value is highest for licences regarding ML categories which cover major conventional weapons. Jets and frigates are just much more expensive than machine guns. When looking at the number instead of the value of licences, other ML categories make it to the top of the most issued licences. Most licences were, as when looking at the value, issued for ML10 (aircraft) with a number of 1167 for the years 2003 to 2014. On rank two, however, is ML1 which covers small arms with 829 licences being issued worth €0.91 billion. While Belgium issued only 10% of the ML1 licences when looking at the

4_{The construction of the Eurofighter Typhoon is coordinated by the Eurofighter Jagflugzeug Gmbh which consists of} three companies and four countries: BAE Systems (UK), Airbus Group (Germany, Spain) and Finmeccanica (Italy).

(25)

21 number, Belgium accounts for 58% of the total when looking at the value. Another big share of ML1 licences was issued by Germany with 32 when looking at the number of licences and 12% when looking at the value. Small arms are usually put in one group with light weapons (ML2) under the term SALW. These two categories, together with the necessary ammunition (ML35_{), account for a share of 16% of all licences}

issued when looking at the value and 17.8% regarding the number of licences, which is the biggest share regarding the number of licences. The third most licences were issued for ML9 (vessels) with 679 licences (prime supplier France). On rank four is ML18 with 686 licences. Category ML18 covers equipment for the production of products covered by the ML. Germany has issued the biggest share of ML18 licences with a share of 85% when looking at the number of licences issued and 78% when looking at the value of the licences.

Most of the ML18 licences Germany issued are linked to the production of G-36 rifles in Saudi Arabia under licence of the company Heckler & Koch. Licences for this project were issued in 2008 and the first assembled G-36 rifle was finished in early 2011 (Duquet 2014: 57). This project is not the first of its kind. In the 1970s, G-3 rifles and MP-5 sub-machineguns were also manufactured in Saudi Arabia under the licence of Heckler & Koch (ibid.). For the production of the G-36 rifle there are five key components necessary which get exported by Germany (Zeit Online 2015a). Since mid-2014, no licences have been granted for these components which got attributed to stricter arms export controls of Sigmar Gabriel, since 2009 leader of the Sozialdemokratische Partei Deutschlands (SPD) and since December 2013 Vice-Chanelor (Zeit Online 2015b). This not only led to complaints on the part of Saudi Arabia but Heckler & Koch sued the Federal Government for damages in October 2015 (Zeit Online 2015a). The lawsuit is still pending.

6.2 Denied Licences

While 8348 licences were issued between 2003 and 2014, only 31 were denied. In percentages this means that only 0.37% of all licence applications were denied which is not only a fraction of the total licence applications but also less than half of the worldwide yearly average of 0.93%. For licence denials no value is reported but it is possible to look at the average value of licences for the 22 ML categories. Looking at the average value of denied licences, the share of denials is with 0.33% slightly lower than for the number of licences. The value of licences can, however, vary to a great extent also within one ML category which is why only the number of denials will be used here. All the denials were made between 2011 and 2014, with an exception of 2004 where one licence was denied. In 2011 only one licence was denied (0.12%) but the number jumped to twelve (1.27%) in 2012 followed by eleven denials (1.25%) in 2013. For 2014 the number of denials decreased again to six (1.25%). Looking only at the years 2011 to 2014 an average of 0.97% of the licences to Saudi Arabia were denied which is a bit higher than the worldwide yearly average.

5_{ML3 includes ammunition for SALW but also for ML12, which comprises high velocity kinetic energy weapon} systems used to destroy armoured vehicles. Only two licences with a value of €10.3 million were issued for ML12 which is why ML3 is attributed here to SALW, although the value of the ammunition destined for SALW is likely to be a bit smaller.