Plugging in the British: EU justice and home affairs

As Britain and the EU have wrangled over Britain’s departure from the EU and its future relations with the Union, the main focus has been on trade and economic relations. That makes sense: the EU is by far the UK’s largest trading partner; and after Brexit, the UK is likely to be the EU’s second largest trading partner. But there has been far less discussion of the other areas in which EU member-states work together, and how the UK might be able to co-operate with them in the future.

In the trade and economic area, the UK will be both an important market for some member-states and a competitor with them for business in third countries. In non-economic areas, however, there are likely to be many areas in which both sides will want to preserve as much as possible of the existing patterns of co-operation. If law enforcement co-operation breaks down, the only people to benefit will be criminals. If defence co-operation fails, the EU will lose access to the resources of Europe’s strongest military power. If UK and EU foreign policies diverge, both parties will find they have less influence over events. 

The European Council’s April 2017 negotiating guidelines for the withdrawal process implicitly reflect the assumption that co-operation will be easier on issues other than trade: of 28 paragraphs, only one makes brief reference to possible partnerships in “the fight against terrorism and international crime, as well as security, defence and foreign policy”.

Despite their common interests, however, in practice the EU and UK will not find it easy to maintain the current level of integration and co-operation after Brexit. The EU is a rules-based institution; and the rules are designed with the interests of member-states in mind, not those of third countries. For the UK, the most important Brexit slogan was ‘Take back control!’ Even if UK foreign policy objectives almost always correspond with those of the rest of the EU, and will still do so after Brexit, the UK will not want to simply accept policies decided by the EU-27. At the same time, in its negotiating guidelines the EU listed “autonomy as regards its decision-making” as a core principle: the UK will not get a veto over decisions relating to foreign policy, defence or security issues, any more than it will over internal market decisions. 

With this in mind, the Centre for European Reform and the Konrad-Adenauer Stiftung began work in 2017 on a series of workshops and publications to explore existing models of co-operation between the EU and like-minded non-members such as Canada, Norway and the United States in three areas: foreign and development policy; defence co-operation and defence industry; and police and judicial co-operation. The aim was to see what the advantages and disadvantages of each model were for each side, and what lessons the UK might learn from the experience of others. 

Legal and political obstacles will not simply disappear just because both parties realise the importance of a close security partnership.

The first policy brief, on foreign policy and development policy, was published in March.1 The second policy brief, on defence policy and capability development, was released in April.2 This third policy brief looks at police and judicial co-operation. It starts by assessing the EU legal framework for justice and home affairs and the UK’s particular position in that framework. It examines what the British government has said about its future relations with the EU in this area, in particular in two of a series of papers on options for its future partnership with the EU-27 published in September 2017 and May 2018.3 It analyses what the EU is saying, publicly and privately, about the sort of future security relationship it wants with the UK. It looks at the legal and political frameworks of relations between the EU and third countries in the security area, and, in particular, extradition, databases and co-operation with EU law enforcement agencies. It examines what those countries think about the advantages and disadvantages of the approaches they have adopted. And it tries to draw some conclusions about realistic aims for the UK, including during any transition period. 

If all goes according to plan, the EU and UK will finalise the withdrawal agreement by around October 2018; in parallel with this process they will agree on the framework for their future relationship. Once the UK formally leaves the EU on March 29th 2019, there will be a transition period of 21 months (unless the parties agree to extend it), during which the details of the long-term EU-UK relationship are supposed to be negotiated and ideally ratified. 

Police and judicial co-operation in criminal matters is one of the issues in the draft withdrawal deal yet to be agreed. This debunks the oft-cited myth that a Brexit deal on security will be easier to agree than one on trade. Just because both parties realise the importance of a close security partnership does not mean that legal and political obstacles will simply disappear.

EU justice and home affairs 

EU justice and home affairs (JHA) comprises a set of policies designed to help member-states manage the negative side-effects of closer economic integration and the abolition of border controls. As EU countries progressively stopped checking people at the borders between them, and in parallel goods, services and capital moved more freely, both law-abiding Europeans and criminals became increasingly mobile. The free flow of capital made money-laundering easier. The development of the four freedoms also led to more people from different nationalities getting married, having children, entering into contracts and buying property in another country. Meanwhile, migrants and asylum seekers were arriving from outside Europe, looking to settle in different EU countries. 

In response to these trends, the 1999 Amsterdam treaty stated that one of the EU’s main objectives should be “to maintain and develop the Union as an area of freedom, security and justice, in which the free movement of persons is assured in conjunction with appropriate measures with respect to external border controls, asylum, immigration and the prevention and combating of crime.” Even so, member-states were reluctant to allow the EU to extend its competence into such sensitive areas of national sovereignty. Consequently, EU JHA remained largely inter-governmental until 2009, when the Lisbon treaty entered into force, placing this domain under the competence of the EU institutions and the supervision of the European Court of Justice (ECJ). 

Much like the single market, the Area of Freedom, Security and Justice (AFSJ) uses the principle of mutual recognition. In the single market, that means that member-states recognise and accept each other’s lawfully marketed products. In the criminal domain, it implies that national authorities recognise and execute each other’s judicial decisions, on the basis that they trust each other’s judicial systems.4 To promote mutual trust, the EU has harmonised laws and procedures where feasible. The EU Charter of Fundamental Rights (the Charter) facilitates co-operation in this area, by harmonising human rights standards across the EU whenever member-states apply EU law.

Title V of the Treaty on the Functioning of the European Union (TFEU) regulates the AFSJ. It covers a wide range of issues: 

• Judicial co-operation in civil matters (such as what happens to a child of Polish and Italian parents if they divorce).
• Judicial co-operation in criminal matters (for instance, how to extradite a suspected criminal from Britain to Germany).
• Harmonisation of criminal law (ensuring that every member-state punishes terrorist offences). 
• Police co-operation (creating an EU police agency, Europol).
• Asylum and migration policies (what to do with an asylum-seeker when he or she first arrives in Europe). 
• JHA agencies like Eurojust (the EU’s agency for judicial co-operation) and Frontex (the European border and coast guard agency). 
• Border controls (agreeing on a common visa policy and making sure that criminals do not go in and out of the Schengen area undetected). 

The Schengen area

The establishment of a passport-free travel area was the driving force behind the development of the AFSJ. In exchange for abolishing most internal border controls, Schengen countries agreed to work together to protect the area’s external crossings, and to co-operate on judicial matters, law enforcement, migration and security.

Schengen association agreements contain a rolling obligation for Schengen countries to adopt EU laws and ECJ case law.

But while all member-states (except for Denmark, Ireland and the UK, as explained below) are bound by EU measures on police and judicial co-operation, not all of them are part of Schengen. And whereas many EU JHA measures, such as the European Arrest Warrant (EAW), are only open to EU member-states, some non-EU countries are part of the Schengen area (see map 1). This situation is due to Schengen’s peculiar history.

Belgium, France, Germany, Luxembourg and the Netherlands signed the Schengen Convention, designed to remove internal border controls, in June 1985. It eventually entered into force in 1995, with Italy, Portugal and Spain also signing up. The Convention was later extended to include Austria, Denmark, Finland, Greece and Sweden. 

Schengen countries had to come up with a number of ‘compensatory’ laws to ensure that they could remove checks without a corresponding loss of security.5 These measures included a common visa policy, an embryonic migration policy and laws governing the exchange of data. The Convention, and all the laws implementing it are known as the Schengen acquis. The acquis was not part of the EU treaties but rather an inter-governmental agreement. EU member-states decided to integrate the acquis into EU law in 1999, with the Amsterdam treaty. The Schengen acquis was then placed under the supervision of the ECJ, ten years before the Lisbon treaty brought the rest of JHA into the court’s ambit.

Today, all EU countries are part of Schengen except for the UK and Ireland (which have their own border controls in the Common Travel Area, although they participate in some Schengen measures on law enforcement and migration); and Bulgaria, Croatia, Cyprus and Romania, which should join Schengen fully once the EU deems they are ready. Denmark is part of Schengen but applies the acquis as a matter of international, and not EU, law (see section on Denmark below). Iceland, Liechtenstein, Norway and Switzerland are not part of the EU but are in Schengen. This matters for Brexit Britain because EU co-operation with third countries on JHA discriminates between non-EU, Schengen countries, and non-EU, non-Schengen countries. 

The EU has Schengen association agreements with non-EU Schengen members like Norway and Switzerland. These contain a rolling obligation to adopt new EU laws and ECJ case law to ensure coherence across all the areas covered by the agreement. If the ECJ and Norwegian or Swiss courts disagree on the interpretation of one of their agreements with the EU, the agreement will be terminated. Even though both countries have a say, ultimately, it is up to the EU to decide what EU laws need to be transposed into domestic law. If they are not, the EU has the power to scrap the agreement.6  

Britain and EU JHA

The UK has never been a full partner in JHA. At best, other countries have seen it as a slightly annoying but necessary partner, tolerated because of its vast expertise in police and security. At worst, it has seriously irritated others by cherry-picking JHA policies.7

The UK and Ireland have secured an opt-in/opt-out regime for EU JHA. This regime covers two different areas:

1) Title V TFEU 

The UK enjoys an opt-in to measures under Title V, which allows it to choose whether to take part in them, sometimes even after they have been adopted.8 For example, the British government did not opt in to the regulation covering cross-border maintenance payments to children and former partners until after it was adopted, once it had made sure that the regulation met UK needs.9  

2) Schengen 

The UK is not a member of Schengen, but it has the right to participate in some Schengen-related measures (including the Schengen Information System (SIS), a database of stolen identity documents and wanted people). Britain and Ireland participate in most of Schengen’s criminal and policing laws, but not the rules regulating border controls, visas and free movement of travellers. If Britain has opted out of a whole area, like migration, it cannot opt in to any measures linked to that area.10 

Despite its opt-outs, the UK generally enjoys a good reputation in the EU as a reliable security partner.

In 2014, with Theresa May as home secretary, the British government exercised its right to opt out of 130 JHA measures adopted before the Lisbon treaty entered into force. Simultaneously, it announced that, for reasons of national security, it would opt back in to 35 of them, including Europol, Eurojust and the European Arrest Warrant (EAW).

In practice, the UK and Ireland have opted into most measures on civil co-operation; the so-called Dublin regulation, which governs the management of asylum claims in Europe; several instruments on law enforcement and police co-operation, like the EAW; and some EU laws to fight irregular migration.11  

Britain is not the only awkward partner in JHA. Denmark participates in Schengen measures, but applies them as international law, and not EU law. This means that Copenhagen has no voting rights in the Council of Ministers on Schengen matters and that the ECJ has no authority over Denmark unless it has opted in to the acquis. Unlike the UK, Denmark does not participate in JHA measures adopted after the Lisbon treaty entered into force. The Danish government needs to negotiate access to every post-Lisbon JHA law, like the new Europol regulation, almost on the same terms as non-EU countries. 

Despite its opt-outs, the UK generally enjoys a good reputation in the EU as a reliable security partner. The UK has been the driving force behind some of the most important JHA measures, including the EAW, the EU’s counter-terrorism and anti-radicalisation strategies and the Passenger Name Record (PNR) system, used for exchanging information about airline passengers travelling to or from third countries. Paradoxically, the UK has always advocated pan-European solutions to security threats and crime, to plug security gaps and catch cross-border criminals. 

What does Britain want?

To judge from the September 2017 ‘Future Partnership’ paper and the May 2018 slides on a framework for co-operation, in essence, Britain wants to leave the EU but stay in those parts of law enforcement and judicial co-operation it likes on more or less the same terms it enjoys now. From a British perspective, this makes sense: the UK has been picking and choosing between EU justice and home affairs measures for many years.

The government’s paper and slides underline shared security threats like terrorism, cybercrime and irregular migration, and stress the UK’s contribution to EU law enforcement and judicial co-operation.12  

The UK government aspires to have a ‘bespoke’ agreement on JHA, going beyond existing co-operation deals between the EU and third countries, and focusing on three main areas: 

• Access to EU law enforcement databases, such Eurodac (which stores fingerprints of asylum-seekers); the Prüm national databases (storing DNA profiles, fingerprints and vehicle registration data); PNR; and the SIS. 
• EU measures to support practical law enforcement co-operation, such as the European Investigation Order (which allows one member-state to ask another to carry out investigations and gather evidence on its behalf, within a deadline of 90 days); and the EAW. 
• Co-operation through EU agencies like Europol and Eurojust.13 

In exchange, the UK is willing to apply EU data protection rules in full; and would be open to exploring dispute resolution mechanisms, in the shape of an international court, a mediator or another adjucating body. 

Both the government’s future partnership paper and Theresa May’s speech to the Munich Security Conference in February 14 clarify that the UK’s preferred option for its future security relationship is an overarching treaty with the EU, covering favoured areas of co-operation in justice and home affairs, rather than piecemeal agreements with each member-state, or with the EU as a whole. 

The partnership paper and the new slides suggest that the EU and the UK could sign a deal similar to the Schengen countries’ association agreements – whereby the UK would be a sort of ‘associated’ member of EU JHA, but without most of the requirements Schengen members must follow (see section on co-operation models below). 

The government’s concept is certainly ambitious. It assumes that a bespoke agreement in this area is simply a matter of will. It fails to acknowledge the rules and structures underpinning EU law enforcement and judicial co-operation (for example, the government’s paper does not differentiate between Schengen and non-Schengen countries when assessing existing models of co-operation). Moreover, whereas Theresa May and the rest of her government have been loud and clear on what they want, they are less so on what they are prepared to offer: the British government has only recently accepted that it will need to comply with EU privacy rules – but has not explained how it intends to do so, beyond saying it will look for an overarching privacy deal with the EU. The UK has also been ambiguous about the role of the ECJ in this area: at the Munich Security Conference, May said she would be open to some form of ECJ oversight. But neither she, nor her officials, have yet spelled out what they mean by that, much to the frustration of their counterparts in Brussels.  

What will the EU offer?

Over the years, Britain’s ad hoc approach to JHA has faced very little resistance in Brussels. EU member-states were more interested in building the EU’s AFSJ than spending time and political capital arguing with the UK. After Brexit, the UK will face tougher scrutiny.    

For the EU, a guiding principle is to preserve the integrity of JHA co-operation both in and outside of Schengen.

The European Commission outlined its position in January.15 The Commission identifies the same three priority areas of co-operation as the UK. The paper also sets out the EU’s guiding principles for JHA negotiations:

• Protecting the Union’s interests and the autonomy of its decision-making process.
• Making sure that non-EU members cannot have the same rights as EU member-states.
• Preserving Schengen and co-operation deals with non-EU countries, without upsetting the bloc’s security partners.
• Ensuring that Britain complies with EU data protection standards and EU fundamental rights.
• Finding appropriate mechanisms to enforce an EU-UK security deal and solve any disputes which may arise.

For the EU, a guiding principle is to preserve the integrity of JHA co-operation across the board: Brussels has carefully crafted a complex partnership structure which it does not want to upset. While non-EU countries (Schengen and otherwise) would want the EU and UK to co-operate closely on law-enforcement, non-EU Schengen countries may be unhappy if the UK gets a similar deal to the one they enjoy. Schengen is hardly more popular in Berne than in London, and Swiss politicians may have a hard time explaining to their voters the point of being in Schengen if, for instance, an outsider can have access to Schengen databases but still maintain its own border controls.16 Brussels also cares about the position of EU countries: Denmark’s 2017 partnership agreement with Europol, for example, was hard to negotiate, provides fewer benefits than full membership and is only valid for as long as Denmark remains a member of both Schengen and the EU (see section on Europol below). The EU is unlikely to give a non-EU country more privileges than those provided to members of the club.

The EU also wants to ensure that the European Parliament is consulted about security negotiations with the UK. In the past, the Parliament has voted down counter-terrorism arrangements with non-EU countries on the grounds that its concerns had not been taken into account. The EU wants to avoid such an outcome in the Brexit talks. 

As law enforcement and judicial co-operation rely heavily on information exchanges between countries, the EU wants to make sure that the UK complies with the bloc’s stringent privacy standards. For the EU, the easiest way to do that is to replicate the system of ‘adequacy decisions’ that it already has with other third countries. An adequacy decision certifies that a country’s privacy standards are good enough for the EU to transfer the data of European citizens to it. The Commission issues these decisions and reviews them periodically, to ensure that standards have not been lowered. 

While many in the EU appreciate the importance of the UK as a security partner, not everybody is convinced that Britain deserves a bespoke deal. In private, some senior EU officials dismiss the British government’s claim that the UK is a net security contributor. Recent EU evaluations on the UK’s participation in the SIS and Prüm databases, which are not yet public, show that, in some instances, Britain extracts more than it puts into EU law enforcement databases. 

Other third countries and their co-operation with the EU

The EU has built a network of informal and formal co-operation channels with third countries on police and judicial matters. Because much of this co-operation touches upon Schengen, Brussels tends to differentiate between non-EU countries which are Schengen members (like Norway and Switzerland); and non-EU countries which are also not part of Schengen (like Canada and the US). 

This section examines EU law enforcement and security co-operation with four non-EU countries: the US and Canada, which are close security allies and, for the most part, like-minded countries while not being part of the Schengen area; and Norway and Switzerland, which are European, non-EU countries and Schengen members. The section focuses on the UK government’s and the EU’s priority areas for JHA co-operation: extradition; access to law enforcement databases; and association agreements with Europol. The section also reviews the influence that non-EU countries have over EU policy making in the field of JHA.17 

EU-third country agreements on extradition

Since 2004, extradition between EU member-states has been governed by the European Arrest Warrant, which allows member-states to issue warrants requesting a suspect’s surrender within 90 days. The EAW differs from other extradition treaties in four ways.

No extradition treaty in the world allows for as much co-operation between countries as the European Arrest Warrant.

First, under the EAW, member-states should surrender people suspected of one of 32 serious offences, even if what they are accused of is not considered a crime in the country where they are located. Second, the EAW abolished constitutional bans on extraditing a country’s own nationals: thus Germany, for example, can extradite German nationals to other member-states. Third, the EAW does not allow EU countries to refuse extradition on the basis that the crime they are wanted for is regarded as a political rather than a criminal act in the country where they are located (the ‘political exception’). Finally, extradition under the EAW is exceptionally swift: the average time for extraditing a suspect is 15 days for uncontested cases, and 48 for contested ones.

The ECJ cannot issue warrants, but it has an important role in reviewing the application of the EAW agreement. For example, the Irish High Court has recently asked the ECJ to rule on whether Ireland should extradite a Polish national to Poland in view of Warsaw’s recent judicial reforms, as the European Commission considers that these reforms erode the rule of law. The Irish court has also asked the ECJ whether or not Ireland should extradite an Irish national to Britain to serve a sentence that would continue after Brexit.

No extradition treaty in the world allows for as much co-operation between countries as the EAW. 

Norway and Iceland have a multilateral extradition treaty with the EU, with procedures similar to the EAW, although with some important differences.18 The Norway/Iceland treaty took 13 years to negotiate: there have been problems in amending the national laws of some EU countries and of Iceland.19 Ireland and Italy have still not ratified the treaty. 

The Norway/Iceland agreement does not give a role to the ECJ, and the EFTA court, which polices internal market disputes for the non-EU members of the European Economic Area (EEA), including Norway and Iceland, has no jurisdiction over justice and home affairs. Instead, the deal specifies that the parties should establish a ‘mechanism’ ensuring that they stay up to date with each other’s case law. 

It takes 18 months on average to extradite a suspect under the 1957 Council of Europe Convention on extradition.

Extradition between the EU and Switzerland is governed by the 1957 Council of Europe Convention on extradition, a non-EU treaty which regulated extradition in Europe before the EAW entered into force.20 Extraditions under the Convention are not automatic and the state of bilateral relations can influence decisions. It takes 18 months on average to extradite a suspect under the Convention. According to Swiss officials, Switzerland sends around 365 people per year to EU member-states and receives around 250 suspects. Berne extradites suspects faster than other members of the 1957 Convention: the average time for surrender is around six months, during which the Swiss administration has to pay for detention. 

The United States has signed bilateral extradition agreements with 26 EU member-states (all except for Croatia and Slovenia). It also has an extradition agreement with the EU as a whole.21 And to facilitate investigation and prosecution, the EU and the US have signed a mutual legal assistance treaty (MLAT).22 While both deals were concluded in 2003, they only entered into force in 2010. And the scope of these multilateral agreements is limited: the surrender deal is focused on enhancing co-operation across the Atlantic by complementing, and not replacing, bilateral agreements; and procedures under the EU-US MLAT are slow and inefficient. 

Canada has neither an extradition treaty nor an MLAT with the EU. The EU-Canada Strategic Partnership Agreement (SPA), signed in 2016, says that the parties should try to boost “co-operation on mutual legal assistance and extradition based on relevant international agreements”, as well as looking for new ways to make co-operation easier.23 Canada has extradition deals in force with at least 12 EU member-states.24 Canada’s Department of Justice has seconded an expert to Brussels to work with EU member-states on extradition. 

Access of third countries to EU JHA databases

Over the past 20 years, the EU has built a vast array of databases (see Table 1). Every database serves a different purpose, from catching criminals to gathering information on visa applications. Each has one or more different legal bases, depending on its purpose: if one part of a database is used for law enforcement, and another to secure Schengen’s external borders, that means different legal bases. This matters for Britain because its negotiating leverage for retaining access to a Schengen database is not the same as for remaining part of a database containing information on air passengers – as there is no provision in the EU treaties allowing a non-Schengen country to participate in Schengen measures. Table 1 gives an overview of EU law enforcement databases. Table 2 shows whether third countries have access to them and, if so, what kind of access. 

Third country agreements with Europol

Europol has association agreements with many countries (see Table 3). Strategic agreements allow police forces to share general intelligence and technical information (such as threat assessments). Operational agreements also permit the exchange of personal data. But none of the existing agreements give third countries direct access to Europol’s databases, like the Europol Information System.

Norway, Switzerland, the US and Canada all have operational agreements with Europol, and can post liaison officers to the agency. Norway has three liaison officers and Switzerland has four. Switzerland puts more information on Europol’s databases than some EU member-states and is the leading third country contributor.  

The US agreement with Europol is fairly comprehensive. The US has liaison officers from six different agencies stationed at Europol, ranging from the Bureau of Alcohol, Tobacco, Firearms and Explosives to the Secret Service. Europol has senior liaison officers working in Washington. Europol also oversees US implementation of the EU-US Terrorist Finance Tracking Programme. In addition to the operational partnership, the US and Europol have also signed an agreement regulating the exchange of personal data and related information.

Canada’s operational agreement with Europol is more restricted in scope, and, by the government’s own admission, only represents a small part of EU-Canada law enforcement co-operation, much of which happens outside the EU framework and through international forums. Canada has two members of the Royal Canadian Mounted Police at Europol and is negotiating to post a third. 

The four countries are allowed to participate in Europol’s projects and initiatives.28 For example, they can participate in Joint Investigation Teams (a legal tool allowing multinational police teams to work together on an investigation).29 Norway and Switzerland recently joined Europol’s taskforce to combat cyber crime. But third countries have no say over how Europol works. Only EU member-states have a seat on Europol’s Management Board (the agency’s governing body). Since its 2015 referendum decision to pull out of EU JHA, Denmark has retained its seat on the Board, but can no longer vote.

Third countries’ influence over EU JHA policy making

Non-EU Schengen countries have a better chance of influencing the EU’s thinking on justice and home affairs than their non-Schengen counterparts. Norway and Switzerland participate in Council working groups and COREPER meetings (meetings where the member-states’ ambassadors to the EU prepare the work of the Council of Ministers). They also take part in some Commission working groups; and their ministers attend Council meetings on JHA. Schengen countries often find that being in the room matters. As an official from one of the countries concerned put it, “because decisions are often taken by consensus, a seat at the negotiating table is crucial – regardless of whether or not you are allowed to vote.” 

Norway, Switzerland and Iceland often negotiate together, which strengthens their hand. In contrast, Canada and the US do not sit in EU meetings and have to rely on EU ‘insiders’ both to get information and influence Brussels on their behalf. Traditionally, the UK has played that role. After Brexit, non-EU countries will need to rely on other friendly member-states.

Assessment of the existing models of co-operation

Much to the UK government’s dismay, the EU seems determined to apply the principle of ‘differentiation’ (‘no better out than in’: a non-EU member must not have more rights and fewer obligations than a member) in all areas of the Brexit negotiations.30 In the field of JHA, this means that the EU will try to replicate existing models of law enforcement and judicial co-operation with third countries. This is not (only) a question of legal rigidity, as some in the British government like to think, but rather one of both strategy and efficiency: by following tried-and-tested models, the EU is protecting the system’s carefully designed balance between Schengen and non-Schengen members while using its negotiating resources wisely. For its part, the British government thinks that none of the current partnerships fit the UK’s special position as a security partner, and is seeking a bespoke agreement. This section analyses the benefits and shortcomings of existing models of EU-third country co-operation in the three priority areas identified above.

• Extradition

If Britain fails to secure a deal on extradition akin to the European Arrest Warrant after Brexit, it will have three options.

First, Britain could seek bilateral extradition agreements with other European countries (the American and Canadian model). The UK could prioritise partnerships with countries like Poland, from which it receives a particularly high number of warrants. A system of 27 bilateral treaties would comply with one of Britain’s red lines, as it would give no role to the ECJ. But recent case law from the ECJ shows that escaping the Luxembourg court’s jurisdiction may not be that easy, even in cases where bilateral treaties apply.

There is no legal base in the EU treaties for a non-EU, non-Schengen country to access Schengen data.

In April 2018, the ECJ was asked about the extradition to the US of an Italian citizen, Romano Pisciotti, who had been arrested while in transit in Germany.31 After serving his sentence, Pisciotti sued the German government on the basis that, by virtue of his free movement rights, Germany should have treated him as a German citizen and thus not extradited him to the US. The Court dismissed his claim but said that, in cases like this, the home state of the suspect (in this case Italy) should be allowed to issue a European Arrest Warrant and get its national back. The home state should then consider the extradition request from the third country (if they have a bilateral extradition agreement). This ruling complicates the surrender of EU citizens to non-EU countries, as not all have bilateral agreements with all 27 members of the EU, and some surrender agreements may be more generous than others. 

A system of 27 bilateral treaties would be harder to negotiate and less efficient than a single, pan-European extradition treaty. Canada and the US do not have deals with all EU member-states, making it easier for criminals to seek safe havens. If current bilateral treaties are anything to go by, most EU member-states would refuse to extradite their own nationals to the UK. In 2011 Portugal refused to surrender George Wright, a Portuguese citizen convicted of murder in the US.

Second, Britain could fall back on the 1957 Convention (the Swiss model). This would have the advantage that no further negotiation with EU countries would be required (unless the UK wanted to supplement the Convention with additional bilateral agreements with strategic partners). One potential problem, however, is that, as the EAW was supposed to replace the Convention completely, some EU member-states may need to enact new laws to re-implement it vis-à-vis the UK. (The UK itself would need to amend its 2003 Extradition Act). This means that it might take some time for EU countries and Britain to be able to apply the Convention. 

Finally, Britain could seek a surrender agreement similar to the one Norway and Iceland have with the EU (the Norwegian model). Of all the existing models, the Norway/Iceland agreement is the closest to the EAW. The deal works around the problem of judicial oversight by setting up an autonomous dispute resolution mechanism. This makes this model very attractive for the UK. But the Norway/Iceland deal has four main shortcomings: first, it allows any party to refuse to extradite their own nationals. So far, fourteen countries have said they would only extradite their own nationals under certain conditions; and seven member-states will not surrender their own nationals to Iceland and Norway.32 Second, it allows parties to trigger the political exception (nine countries have said they would only surrender people suspected of political crimes under certain conditions).33 Third, it is unclear how the mechanism set up by the surrender agreement would work, who would be part of it and what would happen if it were asked to rule on issues of criminal procedure and fundamental rights, such as whether or not to extradite somebody (as only courts can do this). Finally, the Norway/Iceland agreement took a long time to negotiate and is still not in force.

• Databases

There is no legal base in the EU treaties for a non-EU, non-Schengen country to access Schengen data.

If Britain seeks to retain access to the Schengen Information System, it will have three options.

The first would be to ask Europol, or a friendly EU or Schengen member, to run a search every time UK authorities need information from SIS (the Canadian and American model). British law enforcement authorities would not have direct access to SIS and it would take some time for them to get information. Indirect searches can only yield a ‘hit/no-hit’ answer, so British authorities would know that a person (or a stolen item) is indeed in SIS but would have no further information, unless the ‘owner’ of the data granted Britain access. For this option to happen, the UK would need to conclude bilateral deals with several or all EU member-states so that they could run searches on behalf of the UK and grant British authorities access to further information.

The second option would be to follow the Norwegian and Swiss model. Norway and Switzerland have direct access to SIS on the basis of their Schengen association agreements. The UK would retain direct access to the law enforcement part of SIS but would need to pay a small sum into the EU budget (in 2015 Norway paid €6 million to participate in JHA); accept ECJ supremacy over British courts on issues related to Schengen data; and follow EU privacy standards, including on matters of national security like surveillance. 

Both Schengen and non-Schengen countries can exchange information with Europol and post liaison officers to the agency.

But this model may simply not be on offer: currently, the Schengen Information System is open only to Schengen members and EU countries. So a third option for the British government would be to seek a bespoke agreement which maintained as much of the status quo as possible. For that, the EU and the UK would need to negotiate a new legal base and agree on data protection rules and judicial oversight. 

To keep Eurodac, the UK will have two options: either following the Canadian model, whereby Canada can ask an EU country to run a search for them through Europol; or retaining access in the way Norway and Switzerland do by remaining in the EU’s asylum system (which makes the country where asylum seekers first enter the EU responsible for looking after them). The first option would require bilateral agreements with EU member-states (Canada cannot yet make use of the system as these bilateral agreements are not in place). The second would mean that Britain would remain bound by EU legislation in the field of migration and asylum, including the EU Charter of Fundamental Rights. 

Non-EU, non-Schengen countries do not have access to the Prüm databases. If the UK wants to stay connected, it would again need to follow the Norwegian model (direct access), under the conditions of judicial oversight, budget payments and regulatory alignment for Schengen countries mentioned above. Conversely, as Norway and Switzerland do not have a passenger name records (PNR) agreement with the EU, Britain would be interested in following the American model. The US signed a deal with the EU in 2012 to exchange PNR.34 The EU-US 2012 PNR agreement is the latest in a series of transatlantic treaties to exchange air passenger information, beginning in 2004. Although the Commission’s latest review of the 2012 PNR deal praises transatlantic co-operation in the matter, it took the US a long time to forge an agreement with the EU. A 2004 treaty was annulled by the ECJ at the request of the European Parliament.35 It took three years for the EU and the US to negotiate a new PNR agreement – which still did not entirely satisfy the Commission and the Parliament. The latest 2012 agreement is subject to periodic review by the European Commission. Similarly, the ECJ brought down a 2014 EU-Canada PNR deal, forcing the Canadian government to renegotiate a treaty to share air travel information. 

• Europol

Assuming that the UK will seek the closest partnership possible with Europol, it could follow any of the available models (Norwegian, Swiss, American or Canadian). All of these countries have deals with Europol allowing them to exchange information and to post liaison officers to the agency. But none of these countries has direct access to Europol’s databases, which makes operational co-operation harder. No third country has a seat on the agency’s Management Board, nor are they required to pay into Europol’s budget. 

Britain may also want to consider the Danish model. Denmark’s new partnership with Europol allows Copenhagen to request information from the agency. But Danish police and security services can no longer interrogate databases on their own: only Danish liaison officers can access the Europol Information System, which means that searches take more time. And in exchange for having limited access to Europol’s data, Copenhagen has to pay into the agency’s budget and accept the oversight of the ECJ. 

What sort of relationship should the UK try to get? An EU-UK security treaty

Britain has two-and-a-half years to negotiate a new security relationship with the EU. EU law will apply to Britain during the transition period. The aim of this period is to avoid a ‘cliff-edge’ situation for the UK, and allow the parties enough time to reach an agreement on some of the thorniest Brexit questions. Once this period is over, in December 2020, EU JHA measures will no longer apply to the UK. 

The draft withdrawal agreement also suggests a solution for specific cases of police and judicial co-operation which are continuing at the end of the transitional period. Articles 58 and 59 say that EU laws regulating judicial co-operation on police, law enforcement and criminal matters, as well as exchange of data, should apply to acts initiated before the end of the transition period and not completed by the end of this period (so as of December 31st 2020). So, for example, the EAW would apply to a French request to extradite somebody from Britain issued in November 2020 that  has still not been dealt with by December 31st. But if the surrender order were made on January 1st, 2021, a new extradition deal would apply. The British government has still not agreed to this.  

The major obstacle to an agreement on the  transitional provisions is that Britain’s red lines and its stated ambitions for the future security relationship are incompatible. Initially, the government excluded any role for the ECJ (although this position has somewhat softened in recent months); it does not want to align UK law fully with EU law; and would prefer to pull out of the EU Charter of Fundamental Rights altogether.36 The time for finding a solution is running out. But the British government seems to be unable to move beyond the vague idea of a bespoke agreement, whereas the EU insists that, whatever this means, it is not on offer. 

To speed up negotiations in this area, the EU and the UK should first clarify what their future security partnership will cover; second, they should spell out what both parties would be prepared to give in exchange; and, finally, London and Brussels should agree on the shape of their partnership – whether this is best served by a treaty or stand-alone agreements.

What would the future security partnership cover?

To judge from both parties’ current negotiating positions, the future EU-UK security partnership will focus on three issues.37

The more realistic option for Britain would be to seek a surrender deal similar to the Norway/Iceland-EU agreement.

On extradition, Britain is unlikely to convince its partners to replicate the EAW just for the UK. The biggest problem would be getting countries to lift constitutional bans on extraditing their own nationals, because this would require constitutional changes and even referendums in some EU countries. (Germany and Slovenia, for example, would need to change their constitutions. In Slovenia, constitutional change can trigger a referendum).38 In fact, these constitutional bans will start to apply on Brexit day, in March 2019 – member-states are allowed to refuse to extradite their own nationals to Britain after it formally leaves the EU, according to Article 168 of the withdrawal agreement.

The least damaging and most realistic option for Britain would be then to seek a surrender agreement similar to the one Norway and Iceland have with the EU. But even if the UK can start negotiating a surrender agreement before it leaves the EU in March 2019, inevitably it will be faced with a gap before the new treaty can enter into force – as the European Parliament will need to approve it and EU countries may need to make some changes to their criminal laws. In that interim period it will have to revert to the inefficient 1957 Convention. The question is how long that interim period would last. Apart from time pressure, the biggest problem in negotiating a surrender agreement is likely to be the issue of judicial oversight. 

There are several options to solve this problem, none of which is perfect. First, the UK could try to replicate Norway and Iceland’s dispute resolution mechanism for extradition with the EU. Second, the UK and the EU could devise a totally new EU-UK court with jurisdiction over extradition (and perhaps other EU JHA matters). This court could be built from scratch or be a separate part of the ECJ (such as a panel with jurisdiction over criminal justice, as suggested by Petra Bárd).39 While this would be attractive for the British government, the EU is unlikely to agree with such a court, as it would undermine the integrity of EU law (this court would have jurisdiction over intra-EU warrants). Finally, the UK could agree to accept the oversight of the ECJ over surrender procedures between Britain and the EU-27. 

On Schengen databases, London and Brussels will need to be creative if Britain is to retain access to the SIS, as there is no legal basis for a non-Schengen, non-EU country to do so. It is technically possible to create ways for the UK to stay in the law enforcement part of Schengen (the EU could devise a new system whereby Britain, as a special security partner which had access in the past, could be allowed direct access to SIS). But this would be legally and politically complicated. The EU would need to find a new legal base to keep Britain plugged into Schengen databases, risking alienating Schengen and non-Schengen members alike. As it stands, the more realistic option for the UK is to retain indirect access to SIS through Europol or the national authorities of a Schengen country, as Canada and the US do. 

Negotiating access to non-exclusive and non-Schengen databases should be easier: if Britain wants to retain access to Eurodac, it will probably have to remain a part of the EU’s Dublin asylum system. Staying in Dublin while leaving the EU may seem counter-intuitive, but the UK, as a member of the 1951 Geneva Refugee Convention, is obliged to take in refugees and forbidden to send them back to unsafe countries. It can send them back to safe countries, however. The UK is a net beneficiary of the system (in 2016, the last year for which data are available, the UK sent 553 asylum-seekers to other member-states and received 355), so it should have an interest in staying in, at least as some sort of associate member.40 

It should be fairly straightforward to negotiate associate status for the UK in the EU Passenger Name Record scheme. 

It should be fairly straightforward to negotiate associate status for the UK in the EU PNR scheme. After all, Britain drove the adoption of the system, and it already has all the necessary technical requirements in place. Associate status in the existing EU scheme would be better for the UK than trying to negotiate a separate EU-UK PNR agreement. Although there is no precedent for a non-EU country accessing ECRIS (not even non-EU Schengen countries do), the British government could try to convince its EU counterparts of the added value of having Britain connected to the system, as the UK is the fourth largest user of ECRIS. Finally, the UK could negotiate an agreement similar to the one Norway has to retain access to the Prüm databases. But if Britain wants to retain access to EU JHA databases, it will need to comply with EU data protection standards (see section below).

The UK should try to negotiate a close partnership with Europol, an agency it has done much to shape over the past decade. Britain already has the largest liaison bureau at Europol and will be able to post liaison officers from key agencies and bodies (for example, from HM Revenue and Customs, the National Crime Agency or the Security Service (MI5)). But Britain’s co-operation with Europol would be easier if it could, in addition, retain some positions on the agency’s permanent staff, to facilitate communication between Europol and the British authorities, including on access to information. The UK is, however, unlikely to retain direct access to Europol’s databases, as the EU has denied this option to an EU country (Denmark). Paying into Europol’s budget is not compulsory for third countries (Norway does not), but it would be a sign of good will and could earn Britain a few more perks, especially if London wants to keep British staff stationed at Europol. London would need to chip in enough money at least to support Europol’s operations on, for example, disrupting smuggling networks or dealing with the aftermath of large-scale cyber attacks. Unlike previous partnership deals, any agreement between Europol and the UK would need to be approved by the European Parliament, in line with the new Europol regulation.41 

What would be the price of a security treaty?

There are three main things the EU is likely to insist on before agreeing to a security treaty with the UK.

First, Theresa May and her government would have to accept that some sort of international court will be needed after Brexit – not only to review extradition requests, but also to ensure that Britain complies with EU data protection standards, and to rule on the validity and interpretation of the security treaty. Whether this is an entirely new court or draws from existing tribunals would depend very much on the shape of the treaty and the outcome of negotiations on the wider Brexit deal. In any case, the British government will need to come to terms with the fact that the ECJ, by shaping what EU countries are able to do in relation to the EAW, information sharing and police co-operation, will also influence how any future EU-UK security agreement operates.

Second, if Britain wants to retain access to EU law enforcement databases, it will have to comply with EU privacy standards. This may sound relatively straightforward (after all, EU data protection standards have been part of British law for 20 years now, and the UK has said it will apply the new EU data protection regulation, which came into force on May 25th 2018), but Brussels and London disagree on how to make it happen. 

To justify giving the UK a special status, the EU may demand that London not only retains EU privacy laws, but is also willing to allow the European Commission to scrutinise British data protection standards periodically. The EU may demand to know exactly what London is going to do with the data and with whom it plans to share it. The EU would prefer to do that through an adequacy decision. That would enable the Commission to look at British data protection laws; and also to examine legislation on national security (such as the UK’s ‘Data Retention and Investigatory Powers Act’, DRIPA), which also affects the transfer of EU law enforcement data to and from the UK. Given the EU’s dislike of Britain’s intelligence regulations – the ECJ said in 2016 that parts of DRIPA were illegal – and Brussels’ suspicions of the UK’s ‘special relationship’ with the US in intelligence, the EU will want to be reassured that EU data is always treated in a way it deems compatible with its stringent privacy standards. The European Parliament can ask the Commission to withdraw or amend an adequacy decision at any point. 

Negotiating a security treaty may be easier if Britain decides not to withdraw from the EU Charter of Fundamental Rights.

The British government, for its part, would prefer to have an overarching information sharing agreement with the EU, covering data and information transfers for commercial, law enforcement and even military purposes.42 A similar treaty already exists: the US has negotiated a ‘Privacy Shield’ agreement with the EU for the transfer of commercial data and an ‘Umbrella Agreement’ for the transfer of law enforcement information. The ECJ has the power to review both agreements and did strike down a previous data sharing treaty with the US in 2015.43 The EU has so far refused to open negotiations on an all-encompassing data sharing agreement with Canada, citing privacy concerns. 

Third, the UK government might find negotiations on a bespoke security treaty easier if it were willing not to withdraw from the EU Charter of Fundamental Rights. The Withdrawal Bill, which will transfer most EU statutes into British law after Brexit, specifically rejects the Charter. In April, the House of Lords, Britain’s upper chamber, amended the Withdrawal Bill to include the Charter. But this amendment is likely to be rejected once the Bill returns to the House of Commons, the UK’s lower chamber, which ultimately approves legislation. The rights and freedoms contained in the Charter underpin co-operation in the AFSJ by making sure that all EU member-states adhere to the same human rights standards when applying EU law. The British government is right to say that the UK’s Human Rights Act (which made it easier for people in the UK to assert their rights under the Council of Europe’s European Convention on Human Rights) already offers a high level of protection against human rights abuses in Britain. And, of course, the UK is unlikely to morph into an autocratic regime. But the protections afforded by the Charter go beyond those of the Convention, and are crucial for EU co-operation on matters like extradition and information sharing. All EU countries should follow the principles of the Charter when executing arrest warrants (presumption of innocence, or right not to be tried twice for the same criminal offence, for example). Exchange of data on European citizens should comply with the requirements of Article 8 of the Charter (protection of personal data, right to access and rectify own data and the need for express consent for the gathering of personal information, among others). Retaining access to EU databases or striking a good deal on extradition would be less difficult if Britain decided to stay in the Charter.

What would be the shape of a security treaty?

The UK government wants to negotiate something akin to Norway and Iceland’s Schengen association agreements.44 Technically, this suggestion makes sense: it would comply with both the UK’s government demand for a ‘dynamic’ security partnership, and the EU’s insistence on replicating existing models. But politically, it would be virtually impossible for the EU to put the UK on the same footing as non-EU, Schengen countries. As senior officials put it, the only way Britain could get something similar to a Schengen association agreement would be if London signed up to Schengen. That, of course, will never happen.

It may be wiser, then, for the British government to seek a fresh treaty. This could be a stand-alone agreement or part of the wider arrangement governing EU-UK relations after Brexit. Either way, the agreement would need European Parliament approval. A stand-alone treaty will probably be faster to negotiate, but would carry the risk of the European Parliament voting it down or referring it to the ECJ, as it has done in the past with agreements on data transfers for counter-terrorism purposes between the EU, the US and Canada.

Police and judicial co-operation in criminal matters could be part of the wider Brexit deal, with sections spelling out the future arrangements on extradition, access to databases and UK’s participation in EU JHA agencies like Europol. This could be complemented by a chapter on data protection with separate sections for data transfers for commercial and law enforcement purposes.

If the security treaty was part of the wider Brexit deal, it would make it more difficult for the European Parliament to dismiss it, as that would endanger the entire set of EU-UK agreements. And it would reflect the fact that data protection is important both for trade and security; the deal would be more likely to be sustainable in the longer term if it took account of both economic and law enforcement aspects of privacy and data transfers. 

Of course, the main risk of including security in the wider Brexit negotiations is that it might delay a deal in an area where nobody wants to see a cliff-edge. JHA is not like trade, which creates winners and losers: the only losers from increased co-operation in law enforcement are criminals. But, as negotiations progress, it is less clear that a security treaty will be easier (and faster) to negotiate than a trade agreement. This paper has examined the (many) obstacles the EU and the UK will have to overcome if they are to find a suitable security partnership. And deals on extradition and data exchanges with other third countries have sometimes taken longer to negotiate than trade agreements. One way to mitigate the risk of a cliff-edge would be for the EU to agree to extend the transition period for matters of police and judicial co-operation only. The current withdrawal agreement does not include a mechanism to extend the transition period, as the EU is keen to ensure that the UK does not use transition as an indefinite ‘half-in/half-out’ period. But given that nobody voted for the UK to leave EU police and judicial co-operation, it would be wise to exempt this area from Brexit hard-ball, and allow for some legal flexibility. None of this can happen, though, until the UK clarifies its position on the European Court of Justice and the EU Charter of Fundamental Rights .

Conclusion

Brexit is a challenge for the EU. Instead of building a security relationship with a partner, as it often does, Brussels needs to first untangle an existing arrangement, before it can engineer a new deal. Perhaps unsurprisingly, the EU is doing what it does best: using existing rules to protect carefully crafted compromises between multiple countries. In some ways, it is legally and politically easier for the EU to treat the UK like any other third country than it is to upset this balance in order to accommodate a partner that is already half way out of the door, no matter how important that partner may be.

The UK government, for its part, has its hands tied because Britain’s domestic politics make it impossible for negotiators to clarify London’s position on the European Court of Justice and the EU Charter of Fundamental Rights. As a result, the government is asking for a special deal but is unable to say what it could offer in exchange.

Both are opening positions in a negotiation, and likely to evolve over time. But time is a luxury neither the EU nor the UK has: once the transition period is over, in December 2020, London and Brussels will face a cliff-edge unless they have agreed how to keep Britain closely associated to EU police and judicial co-operation while respecting the UK’s wish to ‘take back control’. 

None of the EU’s security partners have a perfect, all-encompassing relationship with the EU. Non-EU Schengen countries like Norway and Switzerland have better police and judicial co-operation with the EU, but in exchange they have abolished border controls and accepted the jurisdiction of the ECJ. Non-EU, non-Schengen countries like Canada and the US are (relatively) free to do what they want because their co-operation with the EU on extradition and databases is fairly loose. Then again, these countries are not on the EU’s doorstep, and neither has previously been a member of the EU. 

On JHA, as in other parts of the Brexit conundrum, the solution is likely to be a half-way house. The EU and the UK could sign a security deal combining elements from existing models but shaped in a different way. The future EU-UK security partnership could be a stand-alone treaty or part of the wider Brexit deal – the latter would minimise the risk of the European Parliament voting it down. And the partnership could draw on the Norway/Iceland extradition treaty and the US relationship with Europol, while keeping Britain plugged into EU law enforcement databases, whether directly or indirectly. Brussels and London could also negotiate a new data protection agreement covering both commercial data exchanges and law enforcement, to underpin their new security deal. For that, both parties would need to agree on mechanisms to solve disputes and ensure compliance with both the treaty and EU data protection and human rights standards.

Whatever the outcome, however, the UK will lose much of its influence on EU JHA. And that is bad news for everybody except the criminals that stand to benefit from a less stringent cross-border regime, from drug gangs in the Netherlands to Eastern European people smugglers and British crooks looking to revive Spain’s infamous ‘Costa del Crime’.

Camino Mortera-Martinez, research fellow, Centre for European Reform, May 2018

This policy brief is part of a joint project with the Konrad-Adenauer Stiftung Great Britain Office. I am grateful to KAS for their support. I am also grateful to the large number of officials, former officials and academic experts whose views, shared off the record, helped to inform the policy brief.

View press release