Travel Bans in Europe: A Legal Appraisal

By Daniel Thym (University of Konstanz), GLOBALCIT collaborator. This blogpost was originally published on the blog of the Odysseus Academic Network and on Verfassungsblog.

In a joint statement of 12 March 2020, President von der Leyen and President Michel ‘disapproved’ of the decision of the Trump administration to impose a travel ban on persons coming from much of Europe. Less than a week later, the European Council unanimously agreed on sweeping restrictions on travel to the Schengen area. Any ‘non-essential’ movements across the external Schengen borders are suspended. This astonishing about-turn followed a week of political activism across Europe to fight the epidemic, including temporary border controls and far-reaching travel restrictions at the internal borders between several Member States.

It is no coincidence that borders play a prominent role in the fight against the coronavirus, since they can fulfil important symbolic functions transcending the practical effects of more police checks. Donald Trump had always been good at exploiting the discursive and symbolic potential of borders to convey a message of political power and to bolster a national sense of belonging, which can be enhanced in times of crisis, when the population feels insecure and threatened. Similar dynamics are at play when Member States reintroduce border controls in the Schengen area. The staggeringly strict new border regimes reinforce that trend.

The political leadership in Brussels was caught off guard by the resurgence of travel restrictions. After having defended open borders during the first weeks of March, the Commission changed course and recognised the legitimacy of border controls, trying to mitigate their impact through political guidelines on 16 March 2020 – and went even further in proposing a ‘restriction on non-essential travel’ to the EU, which the European Council confirmed one day later, thereby reiterating the broader message of a Europe that protects and defends its sovereignty, which are a sort of moderate equivalent of Trump-style protectionism.

Against the backdrop of these political developments, this blogpost presents a legal assessment in eight steps that concentrates on borders controls and the extraordinary travel restrictions for the internal and the external borders of the Schengen area. It will be argued that unprecedented travel bans and border closures for Union citizens are legally problematic. The Commission and the Member States should strive to establish uniform and proportional practices that enhance legal certainty.

(1) Border Controls do not mean Border Closures

It is essential to differentiate border controls from travel restrictions. While the former are fairly unproblematic in the current situation, the latter raise important legal questions. Notwithstanding the EU’s commitment to ‘offer its citizens an area … without internal frontiers’, physical border controls have become a common phenomenon over the past five years. Even before the current pandemic, six Member States including France, Sweden and Austria had reinstated controls at some segments of their borders before the Czech Republic, DenmarkHungaryPolandEstoniaLithuaniaNorwayGermany and Switzerland followed suit or extended existing controls over the past days, while Slovakia announced medical checks in the border area.

These checks are covered by Articles 25 and 28 auf the Schengen Borders Code Regulation, since the coronavirus epidemic can be considered ‘a serious threat to public policy’ requiring immediate action. To be sure, EU law usually distinguishes between public policy and public health, to which the said provisions of the Borders Code do not refer, but this distinction is arguably blurred when an epidemic critically affects manifold fundamental interests of society. In contrast to previous years, there are little doubts that border controls are legal – a conclusion that does not extend to the travel restrictions discussed below.

In practical terms, border controls imply that one may only cross internal Schengen borders at official border crossing points. It is illegal to use the green border and citizens can be penalised under domestic law for doing so. In Germany, for instance, one may be fined up to €1000, even though Member States may rarely enforce it. In a symbolic move to bolster the feeling of closure, border fences are currently being resurrected along some segments of the Swiss-German border, thereby physically separating my academic hometown of Konstanz from its neighbour. Citizens can no longer walk their dogs through the beautiful parks connecting both cities. That nuisance is negligible, however, compared to the severe travel restrictions.

(2) Member State Practices: European Disunity

Border controls erect a physical control infrastructure, but do not conceptually imply travel restrictions. Most internal border controls do not usually take more than a few seconds for Union citizens, who benefit from an unconditional right to enter any EU state as well as associated countries without having to justify the purpose of their visit. All they need is a passport or ID card in accordance with Article 5 of the Free Movement Directive. It is the critical novelty of the current situation that some Member States have enacted severe limits on free movement.

Draconian restrictions which correspond to an almost complete border closure except for those living in the country have been initiated by the Czech RepublicPolandEstonia and Norway. Exceptions may exist for transit passengers, even though they can be difficult to realise in practice. Baltic citizens were blocked at the German-Polish border, until the Baltic states sent ships to take them home after Poland had denied transit. Norway allows the free movement of all citizens of the Nordic Passport Union, while closing the land border rather hermetically to almost anyone else.

Austria systematically rejects anyone entering from Italy who is not in possession of a health certificate in either German, Italian or English validating a negative test result during the past four days. Only nationals or those residing in Austria are allowed to enter (and sent into quarantine). Transit passengers may pass, even though border guards seem to check that cars have sufficient fuel to leave the country without the need for stops. 

Another group of countries, including GermanySwitzerland and Denmark, has introduced an abstract distinction between (non-)essential travel, which seem to have served as a blueprint for the travel ban at the external Schengen border. Those living or working in the country as well as those in transit are usually allowed to enter, while others are rejected, unless they can demonstrate legitimate reasons. While the Germans apply a flexible formula, Denmark and Switzerland have communicated narrow exceptions, such as serious illness or death of a close family member, a court appearance or, in the Swiss case, family visits for important events, while non-married couples living in two different countries are apparently banned from visiting each other.

Hungary and Lithuania, by contrast, appeared to be comparatively generous: only those coming from a high risk region are rejected at the border irrespective of nationality or must go into domestic quarantine for two weeks – until Hungary embraced austere rules prohibiting transit: Bulgarians and Romanians were blocked at the Austrian-Hungarian border, until authorities agreed to transit during night-time. Who would have thought that countries within the Schengen area would emulate communist East Germany in establishing ‘transit corridors’ for Union citizens?

Finally, several states have initiated severe restrictions on internal mobility without, however, restricting access to the country systematically. Italy, France, Belgium and Spain belong to that group, while all the other Member States mentioned above tend to combine strict external entry regimes with comparatively generous internal rules. That cleavage is problematic.

(3) Public Health and the Limits of Free Movement

While travel bans are a matter of executive discretion and limited judicial review elsewhere, including in the United States, EU law has extracted mobility within the single market and the Schengen area from the arcane sphere of state sovereignty. Union citizens benefit from a constitutional guarantee to cross-border movement whose limitations are subject to legal supervision. That is not to say, crucially, that states cannot resort to extraordinary measures in exceptional times of crisis, but they are not free to do as they please from a legal perspective.

It is beyond doubt that the coronavirus qualifies as a ‘disease with epidemic potential’ and may justify, therefore, travel restrictions under Article 29 of the Free Movement Directive. Nevertheless, these restrictions have to comply with the principle of proportionality and be subject to a review procedure in case of complaints – even though these procedural safeguards do not probably involve a right to be admitted to a country provisionally under Article 31 for those who are rejected at the border.

At closer inspection, the application of the public health standard is less clear-cut than it may seem at first. Does it also cover those without symptoms? The wording and the general scheme of the Directive can support such a generous interpretation given that Article 29 does not, unlike Article 27, limit restrictions to the personal conduct and given that the very idea of fighting ‘diseases with epidemic potential’ usually involves preventive action against the wider population. Restricting mobility can be an end in itself to contain an epidemic – in case of the Spanish lockdown not much differently than for the Czech border closure. 

EU law does not necessarily require a uniform response of all Member States either. Coordination may be warranted politically, but the Court of Justice emphasised early on that ‘the particular circumstances justifying recourse to the concept of public [health] may vary from one country to another’ and that it was ‘necessary in this matter to allow the competent national authorities an area of discretion’ (here, para 18; here, para 22). In times of severe public disturbances, that discretion can be reinforced by the safeguard clauses in Article 72 and in Article 347 TFEU, who can influence the interpretation of and possibly even the deviation from other rules of the supranational legal order (here, para 23).

There are, therefore, good reasons that the public health exception can cover travel bans as a matter of principle. Nevertheless, that need not be the end of the legal analysis. Even if the ‘public health’ exception can justify severe and generalised restrictions on mobility, the related principles of proportionality and coherence may require Member States to adapt their practices to comply with EU law. 

(4) Lack of Policy Coherence

It is established case law of the Court of Justice to insist on policy coherence by treating internal and cross-border situations in a comparable way. When restricting the free movement of Union citizens, a Member State must ‘adopt with respect to the same conduct on the part of its own nationals repressive measures or other genuine and effective measures intended to combat such conduct’ (here, para 8; here, para 34). Judges later developed a generic concept of policy coherence preventing states from laying down rules favouring their nationals (here, para 67; here, para 53).

To be sure, states may legitimately restrict travel from high risk areas, such as Lombardy, if the domestic situation is still comparably safe. Thus, Central and Eastern European states, such as Lithuania or Hungary, which have comparatively few official cases, may justify an asymmetric regime subject to the proportionality requirement discussed below. In most other cases, however, it is difficult to justify severe restrictions to cross-border movements, while domestic mobility continues unlimited. Why should people from Luxembourg be prevented from visiting a dear friend in the German city of Trier, while the friend can travel to Berlin at any time? To be sure, visitors from other countries may often spend more time in public places and some German regions have started restricting domestic mobility as well, especially for tourists, but the pronounced asymmetry between domestic mobility and harsh travel restrictions for Union citizens is legally problematic. 

The practical effects of the asymmetric regime for public health will be limited if the rate of infection across the border does not differ markedly from the domestic situation. In such scenarios, border controls are prime examples of symbolic gesture projecting a sense of security that buttresses feelings of national belonging. It is regrettable that some Member States behave like mediaeval fortresses in the current crisis, especially if they do not limit domestic mobility. 

(5) Distinguishing (Il)Legitimate Travel

It is firmly established that restrictions must be proportionate. Under normal circumstances, the Court of Justice insists on an individualised assessment considering each case separately (here, paras 94-99). Even if we accept that, to fight a disease with an epidemic potential, border control agents can currently resort to generalised criteria with little or no room for individualised balancing (here, paras 89-91; and here, paras 64-68), Member States do not have a carte blanche. Instead, a generalised proportionality assessment may distinguish different categories of persons – mirroring the German, Danish or Swiss practice described earlier and the Commission’s proposal for an external entry ban, which, nevertheless, may establish limits to cross-border movement that are too strict.

For decades, the Court of Justice has insisted that the free movement of workers is a ‘fundamental principle’ and required derogations to be interpreted narrowly (here, para 33). Member State should therefore follow the example of Germany, Denmark and Switzerland and the Commission Guidelines by allowing frontier workers and other economically active persons to enter the country to work there under similar conditions as domestic workers. Of course, individual health checks may be performed, but to hermetically close the border to frontier workers turns a blind eye to over 60 years of successful economic integration in the single market.

Other categories can be developed to find a reasonable middle ground. Transit, for instance, should usually be permitted in line with the Commission Guidelines (and it appears excessive if the Austrian authorities require transit passengers to have enough fuel to cross the country). Moreover, we should recognise that all Union citizens benefit from free movement under Article 21 TFEU even if they do not pursue economic activities. Families, for example, should not be separated indefinitely. This Swiss practice might be a good example in this respect, although it appears problematic to categorically exclude nonmarried couples, especially if the travel restrictions continue over a longer period. An even better solution might be the German reference to an umbrella formula of ‘cogent reasons’ (triftige Gründe) that may cover non-family related visits. Such exceptions give border guards leeway to respond to individual needs.

Tourism, by contrast, cannot usually trump public health, while there may be no legitimate reason to prevent people from relocating to holiday homes in countries without a ‘lockdown’, even if they do not live there habitually (and are not covered by exceptions for people residing abroad as a result). Shopping does not appear essential either, although the overall picture appears more nuanced, if we consider that supermarkets near the Swiss border have greatly benefited from transnational customers over the past years. To exclude visitors that had previously been lured can amount to economic protectionism, which the Court of Justice has, rightly, been critical of (here, para 28). Again, debates about shopping may appear extravagant at a time of countrywide lockdowns, but they partly motivated the German travel restrictions.

On one point there seems to be wide agreement: border controls should not hamper the free circulation of goods. Legally, that is mandatory given that the free movement of goods will usually not endanger public health except in the person of the driver – and is, to the contrary, crucial to minimise negative repercussions of the epidemic for businesses and the supply chains of supermarkets and hospitals. The Commission initiative to establish ‘green lanes’ for swift border crossings by trucks is a pragmatic solution at times when there are dozens of kilometres of traffic backlog, not least at the external borders of Austria and Poland. For the history of EU integration, it nevertheless presents a return to the roots given that a green sticker at windscreens had initiated the abolition of border controls under the Franco-German Treaty of Saarbrucken of 1984. That we revert to long-forgotten practices to safeguard basic free movement guarantees shows how dangerous the current situation is for the European Union.

(6) Demonstrating Legitimate Reasons in Practice

Border checks are an administrative activity that is habitually implemented at short intervals and requires swift judgment. That is why the Member States should strive for a clear set of criteria and administrative formalities that can guide decision-making. At present, travel restrictions seem to be rooted in in executive ordonnances in most Member States, which may be acceptable in an emergency situation. As soon as possible, however, respect for limitations under EU law should ‘be ensured by the details and limitations contained in the framework for the practical exercise of the police powers enjoyed by the Member States’ (here, para 38). The categories discussed above may serve as a point of reference.

In practical terms, it is crucial that citizens know what kind of documentation they need to demonstrate legitimate reasons. Legal certainty commands clear standards and, yet, these guidelines should leave room for a flexibility to comply with ECJ case law ‘that evidence may be adduced by any appropriate means’ (here, para 16; and here, para 53). The German federal police, for instance, invites people to bring written evidence, such as a work contract or marriage certificate, while not excluding other means of prima facie evidence. Newspaper reports indicate that border guards verified reasons in (very) brief oral interviews. Austrian insistence on a specific medical certificate at the Southern border appears overtly formalistic even if we acknowledge that Italy is a top-risk country. The Commission should support the spread of best practices across Europe.

In any case, states should refrain from requiring ex ante authorisation which would establish a de facto visa requirement in violation of Article 5(1) of the Free Movement Directive. The ‘frontier worker certificate’ (Pendlerbescheinigung) the German authorities have initiated seems to respect that provision given that it aims at simplifying mobility and does not preclude other evidence. It should be noted, in particular, that the current version foresees the signature of the employer only instead of prescribing a confirmation by state authorities line with a draft version, which I had criticised in an earlier German intervention. Nevertheless, the reappearance of visa-style certificates is a worrying phenomenon for the Schengen area.

There is a surprisingly long line of cases, in which the Court of Justice emphasised the residual responsibility of Member States for their own nationals (here, para 29; here, paras 34-36; here, paras 21-22; here, para 40). On that basis, Member States may grant nationals a privileged right of return, which should, however, include Union citizens habitually residing in a country in line with the Commission Guidelines and the principle of non-discrimination under Article 18 TFEU. In order not to deprive that guarantee of practical effects, Member States ought to give Union citizens leeway in terms of demonstrating residence, since Union citizens do not receive a (declaratory) residence permit in many Member States. They may, therefore, have little proof of residence. In practical terms, basic knowledge of the local language and an adequate description of the connections to a specific city or the workplace may usually suffice.

(7) Asylum Seekers and other Third Country Nationals 

It is important to understand a basic cleavage underlying entry and residence rules in the European Union: While Union citizens benefit from a constitutional guarantee to free movement, so-called third country nationals have a lesser degree of legal protection. That is not to say, crucially, that they have no rights on their side, but the legal contours of statutory guarantees differ markedly. When it comes to formalities, their situation is easier, since third country nationals will usually have a residence permit, which facilitates the documentation of cogent reasons for legitimate travel to return home or to engage in cross-border economic activities. 

In line with previous comments, tourism alone will rarely qualify as a legitimate reason to enter. Similarly, other motives will usually have a lesser weight given that third country nationals do not benefit from free movement under Article 21 TFEU. In exceptional cases, the human right to family unity under Article 8 ECHR may warrant a different outcome, but ECtHR case law on that matter is notoriously vague and subject to an explicit ‘public safety’ caveat. The situation is different, of course, for those who hold a residence permit already. They should be allowed to enter the state who had issue that permit, also considering that the human ‘right to enter his [or her] country’ under Article 12(4) of the International Covenant on Civil and Political Rights can possibly be interpreted to cover long-term residents (here, para 20).

A special case are asylum seekers who may lodge asylum claims ‘at the border’ under Article 3 of the Asylum Procedures Directive and Article 3 of the Dublin III Regulation and must be allowed to enter temporarily, possibly in the context of a border procedure. Member States are, therefore, under a legal responsibility to continue accepting asylum applications, even though it remains an open question to what extent safeguard clauses under Article 72 TFEU or Article 347, mentioned above, may justify exemptions in times of serious threats to public policy, in particular when it comes to secondary movements and rejection at the border that does not violate the prohibition of refoulement. The issue was hotly debated in Germany and Austria during 2015/16 and it is a sign of political realism that the issue is not being raised at the moment. The official Swiss position remains ambivalent insofar as applications at the border are concerned, while the German government has refrained from addressing the theme so far.

(8) Closing the External Schengen Borders

In a spectacular move, the European Council agreed to introduce a pan-European travel ban for the Schengen area as a whole at a videoconference on 17 March 2020. No formal rules were adopted; instead, a simple Commission Communication with no more than 3 ½ pages of written text is meant to guide Member States in implementing the travel ban. There is, therefore, no legally binding legislative act or executive regulation underlying the travel ban, which, rather, emanates from the administrative practices of the Member States on the basis of the Schengen Borders Code, which the Commission Communication aims at coordinating politically.

More specifically, Article 6(1)(e) of the Schengen Borders Code Regulation may serve as a legal basis for a refusal of entry under Article 13 if we conclude, in line with previous comments, that the coronavirus epidemic constitutes a threat to public health as defined in Article 2 No. 21. Unfortunately, the Commission remained silent on the procedure to be followed. In particular, it does not remind Member States that Article 13 of the Schengen Borders Code requires a substantiated decision in line with the standard form attached to the Regulation as well as a right to appeal, albeit without suspensive effect. The Commission should clarify the situation as soon as possible, also indicating to what extent it thinks that either Article 72 or 347 TFEU informs a generous interpretation of existing laws or possibly even a deviation from regular procedural standards. 

Along similar lines, it remains unclear, how the Commission defines the exemptions laid down in the non-binding Communication, such as ‘imperative family reasons’ or ‘persons in need of international protection’, which semantically evades the question whether people coming from potentially safe countries can claim asylum. Moreover, the exemptions are presented in an enumerative manner without indicating whether the reference to ‘non-essential travel’ may be applied in individual cases even if the other criteria are not met – mirroring the ‘cogent reasons’ formula of the German federal police. The Communication was apparently written under utmost time constraints and should be specified as soon as possible.

In doing so, the argument presented above about different categories of (il)legitimate travel may serve as a point of reference, even if we acknowledge that rules for third country nationals can be stricter than for Union citizens. This generic conclusion may extend to the interpretation of the public health exception in line with established case law on third country nationals: similar wording need not mean identical outcome (here, paras 72-74). Just before Christmas, this position was reinforced by the Court in a judgment on Article 6(1)(e) of the Schengen Borders Code (here, paras 31-43; see also here, paras 53-55; and here, paras 40-43). While these arguments sound very academic, they can have tangible repercussions for the implementation of the travel ban at the external border, since they indicate that Member States have more leeway when applying the public health exception in Schengen Borders Code. 

That leeway may influence the definition categories of legitimate travel, can support a generalised proportionality assessment and could vindicate stricter documentation requirements. It may also affect the interpretation of the public health exception at a time when the EU is the epicentre of the epidemic. That entails that the prohibition of inward travel is not only about protecting Europeans, but aims at preventing negative consequences for third states – as the Commission Communication rightly emphasises. It seems to me that it is politically and legally appropriate to take into account such broader effects on world health (here, paras 75-78), even if the Court of Justice came to a different conclusion in a case concerning exit bans for Union citizens (here, para 25). Instead of retreating to national fortresses, Member States should act in solidarity to support each other in fighting the epidemic and minimise the danger of contagion within Europe and beyond.

These diverse considerations show that the Commission Communication can be no more than a starting point of a legal and political process that gradually specifies the contours of the travel ban and supports the spread of best practices among Member States. Otherwise, the Commission and the European Council will not meet the self-declared goal of achieving a uniform application across the Union. Ideally, the EU legislature should stand ready to adopt a Regulation specifying detailed prescriptions as a lex specialis. In doing so, it should clarify the categories of (il)legitimate travel, define procedural standards and elucidate documentation requirements. Swift action can be necessary in times of crisis, but it should not undermine the rule of law in the medium run, on which the European Union depends fundamentally if it wants to achieve its goals.