Intelligence sharing has arguably been one of the most problematic areas of the EU’s counterterrorism efforts. While there appear to have been gradual improvements over time, Europol has certainly struggled to transcend the traditional obstacles to intelligence sharing, and national security and law enforcement agencies are still too often reluctant to share ‘high-grade’, real-time intelligence on terrorism that can be acted on immediately. This is primarily due to the persistence of nationality in international policing and intelligence. Although numerous Council decisions and Commission proposals include an obligation for EU member states to share information, in practical terms, this duty has had little impact because it cannot force member states’ authorities to share more information, that is, intelligence that has not previously been disseminated. This is also confirmed by Europol officials: ‘We know that [national] intelligence services cannot share personal-related and operational-related data with us because of their very strict data protection regimes and there is no use talking it over’ (interview with a Europol official, September 2009).
In either case, information exchange with Europol headquarters is de facto voluntary, and the level of involvement from the various national units varies greatly from one EU member state to another. Consequently, although the intelligence and analysis capacities at Europol headquarters have improved considerably since 9/11, the volume of data that officially reaches Europol remains relatively low. There are several reasons for this, including the defence of sovereignty in matters of ‘national security’, which is further buttressed by a culture of secrecy and the independence of national services, which fear that confidential sources and methods of work could be compromised if intelligence is widely shared. In the case of Europol, these fears are further exacerbated by the fact that prior to 9/11, this EU agency was considered to be a law enforcement support unit only, while after 9/11, it was decided at the political level that Europol would support ‘all competent authorities, including the intelligence services, which were not necessarily ready for this change’ (interview with a Europol official, September 2009). As an alternative explanation, some analysts have suggested that intelligence is also ‘a “currency” to obtain other valuable information or political favors [and therefore] it is not appealing to share it on the basis of general rules with all EU member states’ (Bossong 2008, 19). Others have even argued that national security services may undermine community regimes by submitting low-quality information (Argomaniz 2011, 227).
Furthermore, even when they formally participate in international agencies such as Europol, European police agencies may be reluctant to share information in the absence of several prerequisites, most importantly a sense of mutual trust and a shared expectation of positive outcomes. These, however, cannot simply be created by EU legislation. Moreover, in practice there is a clear preference for bilateral cooperation, which many national agencies consider as ‘the most workable instrument’ from an intelligence perspective (Council of the European Union 2004, 19). As a consequence, ‘Europol represents but an optional bonus, of which the member states can avail themselves at free will’ (Müller-Wille 2004, 26).
An independent study ordered by the European Commission in 2007 provides another explanation for Europol’s shortcomings (John Howell & Co. 2007). The study suggests that the weakness of EU intelligence exchange is due to the presence of an ‘elephant in the room’. The authors recognise that counterterrorism intelligence sharing among member states takes place along two axes. The first axis consists of institutional actors, namely law enforcement, internal security, foreign intelligence agencies and policymakers. The second axis is geographical: global, regional and bilateral (including small-scale multilateral). The authors of the report claim that ‘one of the bilateral players, the USA, is so significant as to represent a separate class of interaction’ (John Howell & Co. 2007, 36). This is because the US (a) has an outreach policy in counterterrorism, (b) actively engages EU member states on policy formulation and implementation, (c) provides technical assistance in the form of training and equipment, and (d) exchanges data. As such, the US ‘is a de facto intelligence hub to which most MS [member states] are in some way connected’, and the ‘EU arrangements risk being crowded out by these relationships’ (John Howell & Co. 2007, 37).
Finally, while formally supporting political initiatives at the EU level, many EU member states simultaneously participate in numerous informal, practitioner-led multilateral networks (e.g. Club de Berne or the Police Working Group on Terrorism), often at the expense of supporting Europol (Bureš 2012). Because of their flexibility, relative independence from national governments and ability to include a broad range of participants on an equal footing, it is generally assumed that these informal networks ‘are more suitable for tackling governance problems or achieving common goals than more hierarchical and formal strategies’ (Den Boer et al. 2008, 103). Especially among professionals, informal horizontal cooperation arrangements are regarded as highly successful, pragmatic and flexible. In this light, it is hardly surprising that a number of studies concerning intelligence sharing consider the preference of EU member states for bilateral and informal multilateral arrangements to be natural and warn against hasty attempts to build supranational intelligence institutions (Benjamin 2005, 15; Müller-Wille2004, 35–6).