Abstract
In 2011, the growing number of pirate attacks prompted several flag states to authorise the use of armed guards aboard vessels. Despite facing the same threat, the United Kingdom, the Netherlands and Italy have adopted three distinct approaches to securing their merchant ships, ranging from the exclusive use of private security companies (PSCs) to the employment of military personnel only. This article conducts a congruence testing of the main theoretical explanations for the use of PSCs on land against UK, Dutch and Italian vessel protection policies. By relying on sequencing as a technique for theoretical synthesis, we develop a multicausal explanation of states’ vessel protection arrangements, showing the varying influence of functionalist, ideational, organisational and political drivers of security privatisation at different phases of the policy process.
Keywords
By the end of the first decade of the twenty-first century, the sharp increase in pirate attacks in the Western Indian Ocean raised strong concerns within the shipping industry, which in 2011 abandoned its traditional opposition against the boarding of armed personnel on commercial vessels. In the wake of this change, a large number of flag states have authorised the deployment of armed guards, either military or private, aboard merchant ships. The majority of European states, including Germany, Greece, the United Kingdom and Scandinavian countries, allowed for the boarding of private security guards. Few other countries, such as the Netherlands, adopted vessel protection policies based on the boarding of military personnel only. Other states, such as Belgium and Italy, fell in between these two extremes, adopting a dual approach that allows for the use of either military personnel or private guards aboard merchant ships.
This article examines the rationale underlying the choosing of different vessel protection policies by conducting a controlled comparison of the British, Dutch and Italian cases. The United Kingdom, the Netherlands and Italy possess comparable merchant fleets, which has resulted in a similar exposure to the threat of pirate attacks. Still, they adopted three different approaches to vessel protection, ranging from the exclusive use of private security companies (PSCs) to the employment of military vessel protection detachments (VPDs) only. Consequently, this case selection allows for both maximising variance in the dependent variable and controlling a key factor shaping states’ security policies, namely, their varying exposure to external threats.
Controlled comparisons have been identified as crucial to assess the explanatory power of different theories. 1 Within-case observation is primarily based on process tracing, identified as the key method for investigating whether the timing and evolution of certain policy decisions coincide with prior, theoretically derived expectations. 2 By tracing the policy processes underlying the evolution of vessel protection policy in three different countries, our study offers a systematic congruence testing 3 of existing explanations of security privatisation and their applicability to vessel protection. We rely on the technique of sequencing 4 to combine existing theories into a thick, multicausal explanation of vessel protection, assessing the relative importance of each of the factors identified by the existing literature as drivers of security privatisation at each stage of the decision-making process.
The evidence collected in this study is based on document analysis and a set of semi-structured interviews with government officials, military officers and representatives of shipping and private security industry associations conducted between May 2014 and February 2017. Due to the sensitivity of the subject matter, we accepted most of our interviewees’ request to remain anonymous as recommended by the scholarship on the ethics of qualitative research. 5 The role performed by each interviewee and the date when the interview was conducted, however, are indicated in endnotes.
While the causes of the increasing use of PSCs on land have received considerable attention, 6 there is little comparative research focusing specifically on the factors underlying the increasing involvement of the private sector in merchant vessel protection. By developing a thick, multicausal explanation of vessel protection decisions and showing the interplay of functionalist, ideationalist, organisational and political factors in informing the choice to resort to either PSCs or VPDs, this article provides an important contribution to private and maritime security studies. Vessel protection decisions are not only foreign policy choices of intrinsic importance but also provide broader insight into contemporary shifts in the management of violence and the institution of sovereignty. Owing to the dual nature of merchant vessels – both fragments of state territory and private assets owned by ship owners – the policies undertaken to protect the shipping industry from piracy offer an ideal example of the growing hybridisation between public and private that is increasingly characterising contemporary global governance. Moreover, private actors are currently playing a growing role not only in combatting maritime crime but also in other maritime security issues, such as sovereign claims over islands in the South Chinese Sea and the provision of migrant rescuing in the Mediterranean. Consequently, the study of governments’ reliance on maritime non-state actors is of relevance not only for security studies but also for international relations (IR) at large.
The article is structured as follows. The next section briefly summarises the four main theoretical frameworks used to explain privatisation of security on land, conceptualising vessel protection policies as a form of sequential decision-making in order to develop a multicausal explanation of the use of PSCs or VPDs at sea. Section ‘British, Dutch and Italian vessel protection policies’ examines the processes leading to the adoption of different vessel protection policies in the United Kingdom, the Netherlands and Italy respectively. Section ‘Explaining vessel protection policies’ assesses the explanatory potential of the existing theoretical explanations of security privatisation against the three selected cases. Section ‘Conclusion’ recaps the main findings of this article, outlining avenues for future research.
Theoretical framework: the drivers of security privatisation examined
The scholarship on the commercialisation of security has elaborated four main explanations for the growing role of commercial actors in state national security, accounting for the increasing resort to PSCs on land on the grounds of functionalist, ideological, organisational and political lines of arguments. Functionalist explanations conceptualise security privatisation as a purposive response to new technological, operational and financial imperatives, providing military organisations under growing financial and personnel strain with the possibility to increase their effectiveness and reduce their costs by outsourcing support tasks to private contractors. 7 This line of argument may also apply to the involvement of private security providers on ships. Due to the high costs and limited number of military personnel available, PSCs may provide a more effective and financially viable response to ship owners’ need for protection.
A large part of the existing scholarship has explained variance in the privatisation of security and military support across countries on the grounds of ideational factors, arguing that defence outsourcing has gained momentum due to the emerging belief in the superiority of market solutions and the commitment to reduce the size and functions of the public sector informed by the rise of neoliberalism. 8 Krahmann has emphasised the importance of theories of the social contract in shaping states’ approach to controlling the use of military force, noting that countries with a ‘republican’ approach to civil–military relations, such as Germany, have displayed a lower propensity to outsource military tasks than states with a ‘liberal’ model of civil–military relations like the United States or Britain. 9 Different market and political cultures, or the normative preference for preserving a state monopoly of violence and refraining from relying on PSCs – sometimes still associated with ‘mercenaries’ and their attached social stigma 10 – may provide insights into explaining why certain states have been more or less open to the involvement of armed contractors in vessel protection. 11
Recent scholarship has emphasised the importance of taking into account the organisational preferences of the main actors involved in the policy process. The outsourcing of diplomatic security, for instance, has provided foreign ministries with the possibility to rely on under-command security providers capable of performing based on their employers’ preferred standard operating procedures, allowing military organisations to resist against the detachment of their personnel for assignments seen as peripheral to their core mission. 12 The different vessel protection approaches adopted by states worldwide can also be explained by the varying organisational cultures and interests of Navies, armed forces and defence bureaucracies at large.
Finally, various scholars have highlighted the political rather than financial and strategic convenience of outsourcing, arguing that military privatisation erodes democratic control over the use of force, creating avenues for the executive branch to circumvent domestic political constraints and reduce the perceived electoral costs of military operations. 13 The use of PSCs at sea may also allow for reducing the political constraints and externalities associated with the deployment of military personnel onboard vessels, downplaying the perceived costs of combatting piracy and avoiding diplomatic complications in case of incidents.
IR theory has been forcefully criticised as unable to fully capitalise on the insights into decision-making processes developed by Foreign Policy Analysis (FPA) despite IR scholars’ widespread acknowledgement that domestic factors play a key role in international politics. 14 The field of private security studies is no exception. While existing research has stressed the need for multicausal explanations for the use of PSCs, most private security scholarship has remained silent on how these different arguments can be integrated, nor has it systematically pinned down the relative importance of functionalist, ideationalist, organisational and political considerations in shaping the stance of the main actors involved in the foreign policy decision-making process. As we show in this article, insights developed by FPA can provide greater rigour to private security studies, helping combine existing theoretical explanations of security privatisation decisions.
Decision-makers often have to engage in a series of decisions regarding a certain foreign policy issue over a prolonged period of time. 15 Consequently, interrelated, sequential decisions are made and reassessed by different actors, each of them with competing interests and priorities. 16 In order to better examine sequential decisions, scholars have long broken down the policy process into different stages or phases. While foreign policy analysts disagree on the exact number and label of each of these phases, these usually consist in (1) the identification of the problem, (2) the evaluation of available policy options, (3) the choosing of a preferred solution, (4) the implementation of this solution, (5) the (re)evaluation of the chosen policy and possibly (6) the making of a follow-on, alternative decision. 17 Although some criticise sequential understandings of foreign policy as in danger of oversimplifying foreign policy as a juxtaposition of semi-insulated phases, there is agreement that conceptualising foreign policy as a sequential process is an analytically useful artifice for a fine-grained examination of complex decisions. 18
Vessel protection policies too, which involve the provision of armed escorts on vessels for a period of several years, can be conceptualised as a sequential type of decision-making. Given the sequential nature of vessel protection decisions, the technique of sequencing first proposed by Jupille et al. provides an especially useful tool for combining existing theoretical explanations of security privatisation. 19 Sequencing allows for theoretical synthesis by showing how variables from different theoretical approaches ‘work together over time to fully explain a given domain’. 20 As the ensuing sections illustrate, existing explanations of security privatisation should be considered as complementary rather than mutually exclusive. Functionalist, ideational, organisational and political factors all play a role in explaining vessel protection decisions. The influence of these factors, however, varies at different stages of the policy-making process. Vessel protection decisions were initially informed primarily by ideational and to a lesser extent by functionalist factors, which played a prominent role in the identification of either PSCs or VPDs as an appropriate solution to the problem of piracy. Once the risks and hurdles of deploying military personnel on merchant vessels became apparent to decision-makers, however, organisational and political considerations gained momentum, shaping a prominent role in the reassessment of existing military solutions and the making of follow-on decisions.
British, Dutch and Italian vessel protection policies
This section analyses the formulation and evolution of British, Dutch and Italian vessel protection policies. In order to clarify the scope of our analysis and justify our case selection, the first subsection provides an overview of the merchant fleets of each of the countries examined and their exposure to the threats of pirate attacks and of maritime crime. The remaining three subsections turn to the policy processes that have informed vessel protection in each of the three cases.
The United Kingdom, the Netherlands and Italy as flag states
In order to trace the processes that have shaped British, Dutch and Italian anti-piracy policies, it is first necessary to look at the characteristics of their merchant fleets, which inform their relative exposure to the piracy threat. Before doing this, two preliminary clarifications are required. First, the figures we use refer to vessels flying the flag of the examined country, without taking into account where the shipping company is headquartered. This choice is justified by the fact that vessels are considered part of flag states’ territory. Hence, it is the flag that determines which states’ jurisdiction, laws and regulations apply on vessels. Second, while the term ‘piracy’ is used throughout the article for the sake of brevity, the threat analysed includes both piracy in a proper sense, which only takes place in international waters, and armed robbery at sea, which occurs in countries’ territorial waters. 21
The statistics published by the UN Conference on Trade and Development (UNCTAD) offer a comparative snapshot of British, Dutch and Italian fleet, highlighting their similarity in size and number of vessels. As illustrated in Tables 1 and 2, the British and Italian fleets have an almost identical deadweight tonnage (DWT), amounting to 17.6 and 17.1 million tons, respectively. While the Dutch fleet has a significantly smaller DWT when compared to the other two, it is remarkably similar to its Italian counterpart in terms of numbers of vessels. As of 1 January 2015, the vessels flying the Italian and Dutch flag were, respectively, 1418 and 1412. The United Kingdom, by contrast, has a slightly larger number of vessels, amounting to 1865 in total. These similarities have been consistent from 2008 to the present, which means they can be assumed as constant during the entire period examined in our study. 22
Rankings of European merchant fleets per DWT, 2015.
Source: Adapted from UN Conference on Trade and Development (UNCTAD), Review of Maritime Transport, 2015.
DWT: deadweight tonnage.
Rankings of European merchant fleets per number of vessels, 2015.
Source: Adapted from UN Conference on Trade and Development (UNCTAD), Review of Maritime Transport, 2015.
Cases of piracy and armed robbery against merchant vessels already increased in the late 1990s, peaking between 2009 and 2011, when they started to decline in absolute terms. Three areas have been particularly affected: South-East Asia (most notably the strait of Malacca); the Gulf of Guinea; and the area between the Gulf of Aden, the Horn of Africa and the Bengal peninsula. This last region – were piracy peaked between 2009 and 2012 – was labelled by the International Maritime Bureau (IMB) as the ‘high risk area’. Table 3, which relies on figures published by the IMB, summarises the number of attacks suffered by European merchant fleets between 2008 and 2014. Data pertaining exclusively to Britain, the Netherlands and Italy are presented in Figure 1.
Attacks against European merchant vessels, 2008–2014.
Source: Adapted from International Maritime Bureau (IMB), Piracy and Armed Robbery against Ships, 2013, 2014 and 2015.
Attacks against British, Dutch and Italian merchant vessels, 2008–2014.
In February 2009, the IMB reacted to the growing piracy threat by launching its Best Management Practices (BMPs), a set of practical guidelines directed to the owners, masters and crew of merchant vessels transiting through dangerous waters. From their third iteration (June 2010) onwards, the BMPs include an explicit definition of the high-risk area, comprising the region between the Horn of Africa and the Bengal peninsula. Up to their third edition, the BMPs only suggested the resort to passive measures, warning against the use of weapons onboard. By the beginning of 2011, however, in the wake of an exceptional rise of piracy (in 2010, 139 attacks were recorded offshore Somalia alone), the shipping industry changed its approach. In February 2011, the International Chamber of Shipping opened to the use of armed guards, a change later acknowledged by the International Maritime Organization (IMO) Maritime Security Committee. 23 The fourth and last edition of the BMPs, released in August 2011, reflects this new approach, no longer discouraging the use of armed personnel for the protection of merchant ships.
The changing stance of the international shipping industry was key to encouraging the use of armed guards on ships. Diplomatic controversies between flag states and coastal states over the presence of armed personnel incentivised the widespread resort to PSCs rather than VPDs as providers of vessel protection. Most notably, in February 2012, two members of an Italian VPD team were arrested by Indian authorities and charged with the killing of two Indian fishermen. The detention of two Italian active duty military personnel caused a long diplomatic friction between Italy and India, which remained in sharp disagreement on the responsibility of the Italian Marines and the jurisdiction under which they should be prosecuted. 24 At the moment of writing, the case is still handled by the International Court of Arbitration in The Hague. 25 In the same period, several private security guards working in an anti-piracy capacity had been arrested by both Senegalese and Indian law enforcement authorities. Such cases, however, did not cause any serious diplomatic confrontation.
Despite the strong incentives to rely on PSCs, not all states worldwide have outsourced the protection of their merchant fleet to commercial actors. As illustrated by the next three subsections, even three European countries possessing merchant fleets of similar size and facing a similar exposure to the threat of piracy like Britain, the Netherlands and Italy initially devised radically different vessel protection policies.
The United Kingdom: a commercial approach
The liberal beliefs underlying British political culture have created a permissive environment for the proliferation of PSCs and their growing involvement in national security. 26 The British private security industry is the second largest in the world (the first in the maritime sector) and includes giants like G4S, the biggest security conglomerate active today. The openness of British decision-makers towards the involvement of commercial actors in security is apparent in their approach to the regulation of the sector. Unlike their US counterparts, British PSCs do not need specific authorisation to operate abroad. The British government has opted for a self-regulatory approach based solely on the raising of standard within the industry by means of voluntary codes of conduct monitored by third parties (e.g. the International Code of Conduct and the American National Standards Istitute/American Society for Industrial Security standards) and the use of government contracts as a way to incentivise PSCs to adhere to ethical and professional standards. 27
The British merchant fleet is the fourth in Europe as of the number of vessels and the fifth in terms of DWT (see Tables 1 and 2). The maritime industry contribution to the national economy is estimated to be around 32 billion GBP per year, amounting to about 2.1 per cent of the national gross domestic product (GDP). About 95 per cent of import and exports transit via shipping. 28 These figures, together with the severe exposure of the British fleet to episodes of piracy (see Table 3), prompted the British government to rapidly enact several measures aimed at contrasting maritime crime. Consequently, the United Kingdom contributed to all the main multilateral anti-piracy missions in the Western Indian Ocean, namely, European Union Naval Force Operation (EUNAVFOR) ‘Atalanta’, North Atlantic Treaty Organization (NATO) ‘Ocean Shield’ and Combined Task Force (CTF)-151, deploying several assets in support of each operation and assuming command of both the EU and NATO missions. 29 The Royal Navy also runs the UK Maritime Trade Operation (UKMTO) by operating the Dubai-based reporting centre tasked with enabling ‘fast communication between merchant shipping and naval forces in the event of an attack’. 30
Given the traditional British openness to the private security industry, it may seem unsurprising that the British government was among the first in Europe in identifying the use of PSCs’ onboard merchant vessel, an appropriate solution to the threat of piracy, preferring this option to a military alternative. This policy was enacted in December 2011 through the Interim Guidance to UK Flagged Shipping on the Use of Armed Guards to Defend against the Threat of Piracy in Exceptional Circumstance. The adoption of a commercial solution, however, was not entirely uncontroversial. In reporting before the Parliament in June 2011, Mark Brownrigg, General Director of the British Chamber of Shipping, stated that ship owners had a preference for having armed protection performed by military personnel and were willing to cover at least part of the costs of their deployment, but were aware that personnel and military constraints made the deployment of UK soldiers aboard vessels unlikely. 31 The British maritime industry has been cognisant that, military missions aside, the fight against piracy was not seen as a priority by the Royal Navy. 32 As stated by Major General Buster Howes (back then commander of EU naval forces in Somalia) before the House of Commons in June 2011, the protection of individual vessels was primarily a responsibility of ship owners rather than a British Navy mission. 33
The involvement of commercial actors in vessel protection in 2011 shows a distinct continuity with a decision made already in 2004, when the International Ship and Port Security (ISPS) code – a set of measures aimed to enhance ship and port security through the introduction of new security requirements – was introduced. Advocated by the United States after 9/11, the ISPS was adopted by the IMO and incorporated into states’ port security regulations. The UK government, and the Department for Transport (DfT) more specifically, decided to provide ISPS-related training to shipping industry workers through private security contractors instead of British military and law enforcement personnel. 34 While the main role in the regulation of private security on land has traditionally been played by the Foreign Office (in coordination with the Ministry of Defence and the Department for International Development), the rise of maritime private security has brought the Department for Transport to the forefront of maritime security regulatory efforts. 35 Like the Foreign Office, the DfT has displayed a favourable approach to the involvement of commercial actors in the provision of security that has remained unscathed from the early 2000s to the present day, regardless of the party in power. The guidelines on the use of PSCs’ aboard vessels originally released in December 2011 have been updated a few times (the last of which, in December 2015). Such revisions, however, only contain regulatory clarifications and amend the size of the high-risk area, without making any substantial revision to the market-based UK approach to vessel protection. 36
The use of PSCs’ onboard vessels has persisted unaltered from 2011 onwards not only because it was consistent with MoD and DfT preferences but also as it proved to be politically cost-effective. Even when British contractors working as private security providers were involved in incidents, no political controversy or heated media debate followed. For instance, four British guards were arrested in May 2013 after entering Senegalese waters aboard a decommissioned Omani Navy vessel, refitted to conduct pirate interdiction off the coasts of Africa. Upon entering Senegal’s territorial sea, the shipmaster and two crewmembers were convicted for the illegal possession of weapons and other minor charges. 37 In October 2013, another group of six British guards was arrested by Indian authorities. The guards have been operating on a ship used as a floating armoury owned by AdvanFort (a US-based PSC) and flying the Sierra Leone flag, the MV Seaman Ohio. According to Indian authorities, the ship had illegally entered Indian waters at the time of the arrest. The guards were therefore prosecuted for illegal import of weapons and are still detained in India at the time of writing. 38 Even if the Foreign and Commonwealth Office (FCO) provided consular support to British citizens in both cases, their conviction did not lead to any government-to-government confrontation. Had military personnel been involved, on the other hand, diplomatic frictions would have inevitably occurred. The severe diplomatic consequences of the arrest of two Italian Navy personnel by Indian authorities (further explained in the section dedicated to the Italian case) did not go unnoticed by British commentators, who identified in the use of private guards an opportunity to minimise UK government liability in the fight against piracy. 39
The Netherlands: a military approach
Initially considered ‘reluctant to outsource’, 40 the Netherlands has increasingly resorted to commercial providers of security and military support starting from operations in Afghanistan, systematically relying on contractors for both logistics and armed security. 41 In 2007, at least 250 armed contractors guarded the perimeter of Dutch bases. 42 While increasingly reliant on the private security industry for the conduct of military operations abroad, the Netherlands has maintained a strict approach towards PSCs operating on its territory. Under the Arms and Ammunitions Law, private guards are prohibited from using firearms. The use of weapons on Dutch territory remains an exclusive preserve of military and police forces. 43
The Dutch merchant fleet generates about 1.5 billion EUR of revenues per year. 44 The safety of sea trade is considered especially crucial due to the importance to the Dutch economy of the port of Rotterdam, the largest in Europe. The tight connection established between the safety of navigation and Dutch national interests led to the conceptualisation of piracy as a ‘threat to the economic security of the Netherlands’. 45 Consequently, the Dutch Navy contributed to EUNAVFOR ‘Atalanta’, NATO ‘Ocean Shield’ and CTF-151 anti-piracy operations. Before the 2009–2011 peak of attacks, both the government and the maritime industry were confident that the abovementioned military missions, in combination with the BMPs, would suffice in containing the threat posed by piracy. Still in Autumn 2009, the worst year per number of attacks directed against Dutch vessels, defence ministry van Middelkoop stated that even if possible, the use of military personnel aboard merchant ships was not advisable because of legal and logistical problems. 46
In the wake of the growing number of attacks, however, the maritime sector changed its stance, starting to forcefully lobby in favour of armed protection. In 2010, the Royal Dutch Navigation Society (Koninklijke Vereniging Nederlandse Reders, KVNR) and the maritime workers union Nautilus International began to advocate for the use of armed guards. According to the Dutch shipping industry, piracy was responsible for the drastic diminution of traffic across the Gulf of Aden, decreased from about 450 daily in the mid-2000s to about 250–300 in 2009. 47 Between 2005 and 2010, 29 ships flying the Dutch flag reported an attack. Only two of these attacks (one of which occurred off the coast of the Horn of Africa), however, resulted in the hijacking of the ship. 48 Although the overall number of attacks remained limited, the industry maintained that the cost of protection measures, in combination with the increase in insurance premiums, jeopardised the financial viability of Dutch shipping.
This belief prompted many ship owners to declare their willingness to take extreme measures. Some companies, such as Jumbo Shipping and Vroon Shipping, announced that they would resort to armed guards even if this was not permitted by Dutch law. In the same vein, in August 2012, the Dutch Royal Dutch Shipowners Association (KNRV) added on this by announcing that it would not allow its members to make transits through the Gulf of Aden unless they employed private guards onboard, regardless of whether this was lawful or not. Other ship owners reflagged their vessel to be able to resort to PSCs. In late 2010, 60 formerly Dutch-registered vessels reflagged. 49
These developments, combined with the change of stance about the use of armed guards made by the international shipping industry in 2011 and the use of guards onboard by several other European countries, prompted the Dutch government to take action. The government identified four different options to the problem posed by piracy: military vessel protection teams, private guards, granting the status of reservist to private guards and a hybrid approach combining elements of the previous three options. 50 Starting in March 2011, the Dutch government ultimately opted for the deployment of military VPDs only. 51
The implementation of the policy, however, proved problematic. From March 2011 to November 2012, only 144 requests for VPDs were filed. In the same period, Dutch vessels conducted between 450 and 500 transits across the high-risk area. According to the Dutch maritime industry, the lengthy time required to process applications, the strict criteria imposed by the Dutch government and the limited flexibility of VPDs discouraged ship owners from requiring armed protection. Moreover, the high costs of this solution – an €150,000 lump sum, plus a daily fee of €25,000 (lowered to €5000 since late 2012) – made the resort to VPDs a very expensive choice. Therefore, many ship owners gave up the option of resorting to military protection. 52
Even if many ship owners refrained from filing an application, the number of available VPDs was still insufficient to meet protection requests. Only 63 of the 144 applications filed in the period between March 2011and November 2013 were fulfilled. The small number of VPDs made the Dutch military unable to protect more than two vessels at the same time. 53 Criticism directed against the military solution devised by the government became stronger when PSCs and the influential Clingendael research institute joined the debate, labelling Dutch vessel protection policy as ineffective and claiming that it would inevitably lead either to reflagging or to the illegal use of private guards. 54 Nevertheless, the Dutch government remained firm on its position and refused to reconsider its policy, arguing that VPDs, in conjunction with the BMPs and military missions in the area, provided sufficient security to Dutch merchant vessels. In Spring 2013, the number of military personnel tasked with protecting Dutch vessels was increased to 175. As admitted by the defence minister, this increase was still insufficient to meet ship owners’ requests. 55
Ship owners’ preferences and economic and strategic considerations did not suffice in forcing the Dutch government to reconsider its approach. Eventually, however, the Dutch Ministry of Defence too grew increasingly reluctant to detach its personnel for VPD duties for several reasons. First, the burden associated with the assignment of elite military units to counter-piracy duties became increasingly heavy. Owing to the awareness that the personnel deployed onboard merchant vessels had to operate in complete self-sufficiency, without being able to quickly obtain reinforcements or emergency medical evacuation, the Dutch MoD conceptualised counter-piracy duties as demanding assignments. Consequently, only Special Operation Forces, such as counterterrorism units, were employed for such duties. These personnel, however, are only available in small numbers and often required for those operational assignments to which the Dutch government has given priority, such as Afghanistan and Mali. 56 Manpower strain was aggravated by the large size of Dutch VPDs. The awareness that Dutch VPDs would need to be self-sufficient made the MoD opt for large detachments consisting in no less than nine personnel, comprising at least three personnel permanently on guard duties on a rotation basis. In 2014 and 2015, no deployed VPD teams were smaller than 11 personnel. 57 Due to the harsh living conditions aboard vessel (ill-equipped to host both a crew and large VPD teams) and the sharp decrease in the number of pirate attacks, the performing of vessel protection came to be considered as a menial and unwelcome tasks distracting scarce elite personnel from more important assignments and undermining their morale.
Moreover, as reported by an official from the Dutch MoD, the Dutch government followed the case of the two Italian marines detained in India very closely, fearing that Dutch soldiers too could be involved in similar incidents. 58 While the activities of Dutch VPDs have not caused any major frictions with third countries to date, obtaining permission for their embarkation and disembarkations in Indian Ocean ports took substantial diplomatic efforts, requiring the lengthy negotiation of ad hoc agreements. Furthermore, the deployment of VPDs entailed severe logistical hurdles, sometimes causing smaller incidents in transit countries. For example, a Dutch VPD transiting through Egypt with their weapons triggered a false alarm among Egyptian authorities, causing the closing of an entire wing of Cairo airport. 59
The increasing wariness of the Dutch MoD against the detachment of military personnel on vessel protection duties has been crucial to bring about an attempt to reform Dutch legislation. In Spring 2016, the Dutch government requested parliamentary approval for the establishment of a dual approach based on the French model. In 2014, the French Parliament enacted a law leaving ship owners free to resort to either PSCs or VPDs. According to the Dutch cabinet’s plan, new legislation following France’s dual approach will enter into force in late 2017 or early 2018.
Italy: a hybrid approach
Italy has traditionally been hostile to the privatisation of armed services. The country’s history, marked by challenges to the state monopoly on violence raised by terrorist and criminal organisations, shaped a national discourse wary of outsourcing or fragmenting the security sector. Budgetary and manpower constraints notwithstanding, this national mindset survived the end of the Cold War and persists until the present day. Some changes have been introduced only recently, as epitomised by the increasing outsourcing of military logistics in deployed operations and the sporadic use of armed security contractors to protect Italian Foreign Ministry personnel. 60
The Italian-flagged fleet ranks fifth in Europe in terms of DWT and eighth in number of vessels (see Tables 1 and 2). The maritime sector as a whole accounts for about 2 per cent of the Italian GDP. 61 What makes shipping even more essential for Italy is its vital role in guaranteeing the flow of key commodities in and out of the country, as about 50 per cent of both imports and exports travel by sea, including a major share of national energy supplies. 62 Like other countries, the Italian fleet faced most pirate attacks between 2008 and 2010 (see Table 3). Given the importance of shipping for the Italian economy, it is unsurprising that Italy was the first country to deploy a military frigate offshore Somalia in an anti-piracy mission (Operation ‘Mare Sicuro’ in 2005). Later on, Italy also joined the multilateral missions EUNAVFOR ‘Atalanta’, NATO ‘Ocean Shield’ and CTF-151. Even if 2009 and 2010 were the worst years in terms of pirate attacks suffered by Italian vessels, Italian ship owners remained committed to the BMPs, back then discouraging the use of armed guards onboard. In this phase, the involvement of the private security sector has been confined to the provision of unarmed guards or consultancy services for ship owners. Although several Italian PSCs called for the use of armed guards onboard Italian vessels, their pleas were not heard. 63
A major transformation occurred in July 2011 with a decree, later converted into Law 130/2011, allowing for the use of armed guards onboard Italian merchant vessels. This shift was mainly prompted by the changing stance of Confitarma, the largest Italian ship owners association, which was in turn informed to two factors. First, in 2011, a large number of attacks occurred, resulting in five hijackings. Given the high costs associated with insurance premiums and the release of hijacked vessels (often achieved by paying a ransom), piracy started to be considered as an increasingly grave threat by ship owners. Second, as has been mentioned already, the international shipping industry changed its own position in early 2011 (a change later reflected in the updated edition of the BMPs released in August). In June 2011, Confitarma too started to advocate for the need to embark armed teams on merchant vessels, threatening the possibility of reflagging had adequate measures not been taken soon. 64
The rationale behind the preference for military providers of vessel protection has been explained by Luigi Ramponi, then president of the Senate Defence Standing Committee. 65 At the time, there was an urgency to address the call for protection made by the shipping industry, but the regulatory conditions necessary for the effective involvement of private suppliers of security were not present because no specific provisions allowing for the presence of contractors and storage of arms onboard had been formulated yet. Consequently, the possibility of using PSCs was therefore included in the law to ostensibly accommodate ship owners’ preferences in spite of being de facto impossible due to the lack of all the necessary follow-on regulations.
The same belief can be read in a resolution approved by the Defence Standing Committee of the Senate on October 2012, which mentioned both inadequate professional standards among private security professionals and regulatory concerns as the reasons why it was impossible to immediately resort to commercial security. Law 130/2011 became operational in October 2011 with the deployment of the first VPDs. Due to the lack of the regulations required for the lawful use of PSCs onboard, resorting to the private option was ultimately impossible until October 2013, when these provisions were finally enacted. Hence, while law 130 outlines a hybrid approach on paper, for two full years the use of VPDs remained the sole option actually available to ship owners. Only in the autumn of 2013 – after the diplomatic controversy with India (see paragraph below) forced the Italian government to reconsider its approach – did the use of PSC become possible. Between 2012 and the autumn of 2014, the large majority of vessel protection missions were conducted by the Italian military. Italian VPDs, consisting of teams from the Italian Navy Infantry Fusiliers Brigade San Marco, conducted over 300 escorting missions, as opposed to the only 50 transits protected by PSCs. 66
In March 2015, the Minister of Defence Roberta Pinotti decided to suspend the use of VPDs. 67 This decision dramatically increased the use of PSCs, which in May 2015 had already conducted no less than 150 transits. 68 The Minister explained her decision based on the decrease in pirate attacks and the implementation of the standards, regulations and training courses ensuring a safe and accountable resort to PSCs. Interviews and document analysis, however, reveal that organisational and political factors played a key role in calling into question the use of VPDs. The Italian Navy was initially supportive of performing VPD duties, seen as consistent with the peacekeeping narrative traditionally used by Italian Armed Forces to win public opinion approval. By late 2013, however, the Navy had refocused its operational priorities on the Mediterranean due to the civil war and NATO intervention in Libya and the ensuing migration crisis. 69 In October 2013, the Italian Navy launched the large Search and Rescue operation Mare Nostrum, which provided a new legitimisation tool for the Italian Navy, forcefully showing its crucial importance in countering a humanitarian and security emergency at Italian Southern borders. 70 The ongoing Italian focus on the central Mediterranean is epitomised by the leading role played by the Italian Navy in anti-smuggling and migration management operations EUNAVFOR ‘Med’.
Political considerations were especially crucial in reshaping the Italian approach, informing the re-evaluation of the existing vessel protection policy and the making of a follow-on decision. In February 2012, two members of an Italian VPD were arrested and charged with the killing of two Indian fishermen. 71 The diplomatic consequences and legal ramifications of the two Italian officers’ trial are beyond the scope of this article. This case, however, forcefully exemplifies the dangers arising from detaching Navy teams on commercial vessels and is considered by the existing literature on maritime private security as the epitome of the political and diplomatic risks associated with the use of VPDs. 72 This awareness clearly emerged during the Italian parliamentary debate. As stated by the Italian defence committee in the autumn of 2012, ‘private security teams could avoid diplomatic complications in case of incidents’. 73
By contrast, considerations, such as cost-effectiveness, do not appear to have played a major role in shaping the Italian vessel protection. Under law 130, VPD personnel remain on the payroll of the Italian MoD. Consequently, only the extra costs arising from their deployment in a vessel protection mission, amounting to €467 per day per person, were to be covered by ship owners. 74 According to a maritime industry representative, this arrangement did not make the resort to VPDs much more expensive for ship owners than the use of PSCs. 75 Nor did this solution impose a heavy financial burden for the Italian MoD, which had to pay salaries to VPD personnel regardless of whether they were assigned to vessel protection duties or not. PSCs, however, are preferred by some ship owners due to their greater flexibility. Tramp vessel especially, which have no fixed schedule and routes, could not be easily protected by VPDs, who operated under a tight schedule and could only be embarked and disembarked in specific ports. 76
Explaining vessel protection policies
As summarised in Table 4, despite facing the same threat, the United Kingdom, the Netherlands and Italy have adopted three distinct approaches to securing their merchant ships. The United Kingdom has immediately adopted vessel protection policies grounded solely on the use of PSCs. The Netherlands, by contrast, has (at least until new legislation is enacted) resorted exclusively to military personnel. Italy has adopted a dual approach that has shifted through three different phases, ranging from the de facto use of military personnel only to the exclusive use of PSCs once the deployment of VPDs got suspended. This section relies on the main theoretical arguments used for the privatisation of security on land to develop a multicausal explanation of British, Dutch and Italian vessel protection decisions.
Vessel protection policies in the United Kingdom, the Netherlands and Italy.
Source: Authors’ elaboration.
Revisions expected in 2018.
Functionalist arguments, explaining the use of PSCs on the grounds of their greater efficiency and cost-effectiveness at a time where military organisations are stretched thin by budgetary and manpower strain, provide some insight into both governments and ship owners’ preferences, but ultimately fail to explain both variance in British, Italian and Dutch vessel protection policies and the evolution of these policies over time. Ship owners’ preferences have not played a crucial role in shaping vessel protection decisions in any of the countries examined. Only in the Netherlands the shipping industry forcefully advocated the use of PSCs due to the high costs and low number of VPDs, resorting to extreme measures, such as reflagging and threatening the unlawful use of private guards. This pressure, however, did not yield any immediate results. British and Italian ship owners, by contrast, were supportive of the deployment of VPDs, willingly accepting to cover part of the costs. Nor were government decisions primarily based on the need to save military budgets and personnel. Only the British government mentioned the military manpower strain caused by multiple operational commitments to justify its unwillingness to assign military personnel to VPD duties. While also confronting demanding operational deployments and severe budget cuts, the Italian and Dutch MoDs did not shy away from detaching scarce military personnel for vessel protection. Hence, as already noted by the scholarship on military privatisation on land, functionalist considerations alone fail to account for variance in the use of PSCs across countries and over time, which can only be explained by taking into account country-specific ideational, political and organisational factors. 77
As noted by several scholars, ideational explanations provide important insights into explaining variance in the use of PSCs across countries. 78 Normative expectations regarding how security on the national territory should be provided and by whom are key to explaining why PSCs were immediately identified as an appropriate solution to the piracy problem in the United Kingdom, but not in Italy and the Netherlands. In the United Kingdom, the ideological commitment to outsourcing part of the state machinery informed by the rise of neoliberalism has certainly contributed to shaping policy-makers’ propensity to involve PSCs in a host of security-related activities, including vessel protection. In the Netherlands and Italy alike, however, the decision to rely on PSCs’ onboard vessels was not primarily shaped by political and economic ideologies. The Netherlands, whose social and economic policies have grown increasingly neoliberal over the past decades, has nevertheless refrained from relying on PSCs’ onboard Dutch ships. The comparatively less neoliberal Italy, by contrast, has been more open to the employment of private guards on merchant vessels.
Moreover, if the decision to resort to PSCs were to be fully explained by decision-makers’ ideological preferences, one would expect such preferences to be consistent over the spectrum of security policies both on land and at sea. Although the existing literature argues that the propensity to use PSCs to protect vessels mirrors their willingness to employ them on land, 79 no such consistency can be found in the three cases we examined. While the Netherlands has relied on PSCs in Afghanistan to a considerable extent, Italy has never resorted to armed contractors in military operations and employed PSCs as providers of diplomatic security only in a couple of exceptional circumstances. Still, Italy has been more open to PSCs’ onboard merchant vessels than the Netherlands.
Ideational explanations focusing on the normative attachment to the state monopoly of violence can partly account for this puzzle. Since ships flying the Dutch flag are part of the territory of the Netherlands, the fact that Dutch PSCs are not allowed to carry arms within Dutch borders is key to explaining the unwillingness to allow for the presence of armed contractors on vessels. On the other hand, Italian legislation allows for the use of armed guards on national territory. This has made it easier for Italian decision-makers to allow PSCs’ onboard merchant vessels, although it took time for this policy to become fully operational due to legal and bureaucratic hurdles. While shedding key insights into differences across countries, however, ideational arguments suffer from indeterminacy in explaining change in the use of PSCs over time on land and at sea alike. Ideational explanations suggest that decision-makers’ preferences – informed by longstanding normative beliefs – should change slowly over time. 80 By contrast, the rapid evolution of Italian vessel protection policies and the ongoing revision of the Dutch approach highlight the importance of other explanations in better accounting for the timing of PSCs involvement in certain nation security policies.
While ideational and to a lesser extent functionalist arguments were crucial in initially identifying either VPDs or PSCs as an appropriate policy solution to the problem of piracy, organisational and ideational factors played a vital role in the reassessment of existing policies and the attempt to devise different, follow-on solutions. Organisational explanations seek to provide greater fine-grainedness into security privatisation decisions by conceptualising military organisations’ interests – in turn shaped by their organisational cultures or role conceptions – as an intervening variable inhibiting or enabling the outsourcing of certain tasks to PSCs. 81 As military organisations have been involved in the formulation and evaluation of anti-piracy policies, looking at the organisational preferences of the British, Dutch and Italian armed forces (and most notably their Navies) provides important insights into vessel protection decisions.
Existing studies of British military culture consider the Royal Navy as an organisation focused on naval and anti-submarine warfare rather than maritime policing and law enforcement. 82 As a combat-oriented organisation with a focus on blue-water operations, the British Navy has actively contributed to military anti-piracy missions but resisted against performing vessel protection and port security roles, seen as outside their core mission. In the Netherlands, the organisational priorities of the MoD and the military personnel assigned to VPD duties played a role in gradually strengthening their unwillingness to perform vessel protection. VPD roles were initially seen as demanding tasks requiring large teams of elite forces. The changing strategic priorities of the Dutch MoD and the preferences of its vessel protection units, who started to consider ship security assignments as menial and unwelcome, contributed to turning vessel protection into a peripheral task distracting scarce elite personnel from more important missions. Italian military role conceptions initially played a key role in encouraging the use of military personnel in the performing of VPD duties. In order to strengthen its legitimacy before Italian public opinion, largely imbued by pacifist and anti-militarist beliefs, the Italian military has consistently presented itself as a constabulary force willing to engage in a range of policing and non-combat operations. 83 Anti-piracy operations and vessel protection resonated with this narrative, and were therefore considered fully compatible with the role conception of the Italian Navy. By late 2012, however, the Navy had refocused its priorities towards the humanitarian and security crisis offshore Libya. Due to the crucial importance of the Mediterranean for Italian national security, maritime piracy, already declining in absolute numbers, was pushed to the sidelines and considered as a peripheral mission that could be outsourced to the private sector.
Military organisations’ cultures, interests and priorities alone, however, do not determine vessel protection decisions. While military preferences are rarely entirely disregarded when planning, implementing and reassessing security policies, national security decisions are ultimately formulated by civilian officials. Explanations emphasising the political convenience of resorting to commercial providers of security play an important role in shaping elected decision-makers’ military mobilisation strategies. 84 Calculations associated with the political rather than merely financial costs of deploying military personnel on vessel protection duties were not key to inform protection policies when regulations about armed guards on merchant ships were first introduced in 2011.
Such considerations, however, became increasingly relevant over time, ultimately playing a crucial role in the reassessment of military approaches to vessel protection and the making of follow-on decisions based on the use of PSCs. The reason is straightforward. Due to the novelty of vessel protection duties, decision-makers initially failed to fully grasp the risks and costs associated with the deployment of military personnel aboard merchant ships. By mid-2012, however, the political risks arising from the deployment of state military personnel on commercial vessels had become apparent due to the ongoing detention of the two Italian Marines, which resulted in an intractable diplomatic controversy with the Indian government. Such an awareness was crucial in informing both the completion of the Italian regulatory framework required for the employment of PSCs in late 2013 and the interruption of VPD deployments in March 2015. Far from being confined to Italy, the spillovers of the incidents also influenced the Dutch debate, strengthening the MoD commitment to having a new law allowing for the use of PSCs onboard approved soon. The awareness that PSCs offered a more politically convenient option also contributed to consolidating the solely private approach devised by the United Kingdom, where the arrest of several private contractors performing vessel protection did not result in any major diplomatic issue and went largely unnoticed by public opinion.
Conclusion
Our sequential congruence testing of the existing explanations of security commercialisation on land provides important insights into why European vessel protection policies vary across countries and over time. In the United Kingdom, functionalist, ideological, political and organisational factors converged in shaping and consolidating a vessel protection approach consisting exclusively in the use of PSCs. Military manpower strain figures prominently in the discourses underlying UK decision-makers’ use of PSCs as vessel protection providers. The propensity to identify purportedly more efficient commercial actors as appropriate providers of maritime security was also informed by Britain’s neoliberal political and economic culture. The organisational culture of the British Navy, focusing on blue-water operations rather than constabulary duties, in combination with the growing awareness that using PSCs would minimise UK state liability in the contrasting of piracy, further contributed to consolidating a solely market-based approach. Italian and Dutch policies have initially been shaped predominantly by a different type of ideational considerations, namely, the normative preference for preserving a state monopoly of violence on national territory. Neither militaries’ budgetary and manpower strain nor ship owners’ vocal preference for cheaper and more flexible private escorts sufficed in overcoming this normative resistance against the use of PSCs, which were not initially identified as an appropriate policy option. The change in Italian vessel protection policies was facilitated by the changing operational priorities of the Italian Navy, which no longer saw anti-piracy assignments as a key mission. In the Netherlands too, the MoD changed its stance primarily because VPDs’ duties were increasingly seen as menial tasks distracting elite military personnel away from the main operational priorities. The decisive factor that brought about a change in the Italian approach, however, were the political considerations triggered by the ongoing diplomatic controversy with India, which prompted Italian decision-makers to identify in the use of PSCs a politically cheaper solution to the need for protecting merchant vessels.
Consistent with other scholarship on private security, 85 our comparative study shows that a fine-grained explanation of vessel protection policies benefits from the use of each of the theoretical arguments identified by the existing literature as the causes of security commercialisation. Existing private security scholarship, however, has not fully specified how these explanations should be integrated and combined. In this article, we relied on sequencing as a tool for theoretical synthesis that allows for a thick, multicausal analysis of the reasons underlying the use of PSCs or VPDs by assessing the varying influence of each of these factors at different stages of the policy process. Specifically, our analysis shows that ideational and, to a lesser degree, functionalist factors have played an important role in the formulation and implementation of vessel protection policies. As already observed by previous scholars, while functionalist reasoning related to effectiveness and cost-efficiency play an important role in privatisation decisions, prior ideational contexts shape governmental actors’ perceptions of appropriate providers of security solutions. 86 By contrast, organisational preferences and above all the growing awareness of the political risks of deploying military personnel on merchant ships have played an increasingly important role in the evaluation and revision of existing approaches, ultimately paving the way to a greater involvement of PSCs in their maritime security policies. Most notably, external shocks like the aftermath of the Italian Marines case may provide the momentum for altering courses of actions that may otherwise be subjected to path-dependency.
Future research on private security should dedicate more extensive attention to the re-evaluation of decisions associated with the use of private or public providers of security in other states and policy domains. As long understood by FPA, the study of change in government decision provides especially important insights into the rationale of foreign policy-making. 87 Moreover, scholarship focusing on the reasons underlying states’ reliance on non-state actors at sea should expand beyond the use of PSCs to protect vessels against piracy to include other topical maritime security issues. In South-East Asia, for instance, China seems to have strategically resorted to maritime commercial assets, such as fishing vessels and oil drilling rigs, to support its sovereign claims over islands in the East and South China Sea. 88 In the Mediterranean, the Italian and Greek Coast Guards have increasingly relied on non-governmental organisations (NGOs) as providers of maritime rescuing offshore Libya and in the Aegean Sea. 89 As our study demonstrated, investigating states’ resort to non-state actors in supporting their maritime security policies provides important insights into contemporary shifts in global governance, thereby contributing to key debates in security studies and international relations at large.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.