Temporary Protection for Ukrainians: learning the lessons of the 1990s?

Forum on the EU Temporary Protection Responses to the Ukraine War 

Contribution by Joanne van Selm, Independent Policy Researcher
11 April 2022

On 4 March 2022, a quite exceptional, revolutionary and original way for Europe to approach arrivals from a major displacement crisis in a neighbouring state was activated. The Council’s decision to implement the EU’s 2001 Temporary Protection Directive gave life to a form of temporary protection which until then had only existed on paper, although the concept of ‘temporary protection’ (or ‘temporary refuge’ and similar terms) has been part of refugee policy for decades.

The initial impact, combined with existing visa free travel, has been to allow Ukraine’s displaced persons, primarily women, children and the elderly, to be welcomed and supported, in stark contrast to European experiences with the majority of refugee arrivals over recent years and decades. It has also, as intended, provided a moment of pause for authorities. Although they need to deal with the practical aspects of sheltering millions of refugee arrivals within a matter of days, they have not had to try to start processing individual asylum applications. From a policy and practical perspective, this has surely been to the immediate advantage of both refugees and the EU states.

Many have pointed to the Directive’s contextual background in the 1990s Balkans conflict and displacement crises, sometimes erroneously suggesting the same form of protection was used then. This contribution will compare current the temporary protection methodology to its 1990s predecessors.

The temporary protection of the 1990s in fact took a vastly different approach than that encapsulated in the Directive. The policy and practical lessons learned from handling displacements resulting from those European conflicts and the intention to do it better if there were to be a next time underpin the broad Directive. In particular, the insight that temporary protection should not be an alternative to refugee status (as it had been cast during the 1990s) but an administrative prelude to the application of the Convention caught the interest of policy makers, and is a major intention of Directive. The form and content of temporary protection being offered to Ukrainians now is conceptually, practically and motivationally quite different from anything offered to former-Yugoslavs, as will be elaborated below. In particular, temporary protection during the Balkan crises could be characterized as reactive, and a reluctant alternative intended to be less generous than asylum. The activation of the directive in 2022 is a proactive measure, welcoming and offering immediate protection and relative stability to displaced Ukrainians. Whereas the temporary protection of the 1990s was focused on return, the focus in 2022 is on supporting the displaced – whether they need to remain long-term, can return or will move on is a matter of later concern.

The major lesson of the 1990s which underlies the 2001 Directive is that Temporary Protection in Europe should be about granting protection to those who seek it, and a strong, organized policy approach that recognizes that asylum systems are not geared up to assess sudden massive numbers of claims. In the Directive, and the practice in 2022, temporary protection no longer means an undercutting of protection standards for people who might well qualify as refugees or with subsidiary protection. Temporary protection, as devised in the Directive, gives breathing space, for governments and for displaced persons, with essential rights guaranteed, while both can assess and ready themselves for how the situation might unfold.

Below, I will address the motivational points of comparison between then and now. After setting out some contextual notes, I will characterize the manifestations of temporary protection through the three phases of the former-Yugoslav displacements, and the Directive as drafted in 2000/2001, to its activation for Ukrainians in 2022. This takes us through the limitations approach to Croats in 1992; the grudging acceptance of, and provisional approach to, Bosnians in 1992-1995; and the emergency stop-gap of 1999, with the mixed arrivals of Kosovars through the Humanitarian Evacuation Programme and those who arrived by their own means. The Directive is pragmatic, if until now hypothetical, an intended win-win for both the refugees and the European states, with beneficiaries granted immediate rights and protections while asylum administration systems are not overwhelmed by sudden significant increases in applications. Due to the brevity of this contribution, the overview will be cursory, with limited illustration of key points. The focus is on policy and approaches, not the legal basis or content of the protection involved.

The major take away is that temporary protection in the EU has changed since the 1990s. The changes are not linked to the altered nature of EU asylum policy, with its underpinning of harmonization and common approaches through agreed directives. Rather they are the result of the understanding that grew out of the 1990s crises of the need for a singular policy tool for a unique type of situation. One could now posit that had the EU updated the Temporary Protection Directive, as had been planned following a 2016 review, doing so without the immediate hindsight of a major influx from a neighbouring or nearby state might have resulted in a tool that was inadequate to the actual situation now faced from Ukraine.

Temporary protection, in the shape of the directive, is exceptional. That in itself is a major contrast with the way in which the concept was handled in the 1990s. This experience implementing the directive should lead to modifications that maintain the EU’s readiness to react to any major situation in which European states are necessarily first responders, and continue to enhance the nature of temporary protection as a policy tool for specific situations, and not as an alternative to Convention or subsidiary refugee protection in the way it was previously applied in the 1990s, but as a complementary measure, alongside the obligations of other Directives.


One major difference, motivationally, is that those fleeing former-Yugoslavia were trying to escape the ‘ethnic cleansing’ of their lands, the fundamental underpinning of the conflicts. European leaders at the time felt this ‘cleansing’ would appear to be supported through the open encouragement of protection elsewhere in Europe or beyond, or through the granting of asylum, which in spite of the 1951 Convention’s cessation clause was typically viewed as a permanent solution.

Those who arrived in EU states were often, as a group, granted a national form of temporary protection. Ukrainians are fleeing an all-out war of aggression, in which civilians appear (early in the conflict, with limited clarity on Russia’s aims, or limits) to be not just indiscriminate victims but targets, and which seems to be intended by the aggressor to bring its victims and their country back into some form of greater Russia, a future against which Ukrainians are fighting. The activation of the 2001 Directive grants blanket protection to all Ukrainians (and some non-citizens who were legal residents of Ukraine).

The situation starting in late 2021 in which Belarussian authorities encouraged displaced Iraqis and others to cross the Polish border has surely also motivated quick, unified and supportive action. That was potentially a precursor to ‘weaponizing’ refugees from Ukraine as a threat to Europe, a possibility which had to be met head on with an open welcome to arrivals who are themselves under threat. No such direct challenge occurred in the 1990s. The Temporary Protection Directive meant the EU had a plan ready for precisely this type of situation. This has, in a turn that will surprise many critics, not only allowed the EU to present a united, humanitarian face to this challenge, but also put it out ahead on refugee protection for Ukrainians while the UK, US and others fumble with accommodating these sudden arrivals, or pressure for resettlement, in less flexible systems.


Throughout the 1990s the European Union was not only dealing with refugee arrivals from former Yugoslavia, but also developing its own very early steps towards harmonization and a Common European Asylum System, and managing the changes and migration-related concerns, indeed fears, of the collapse of Communist power structures in Eastern Europe and the former Soviet Union. There was an internal jockeying for position on asylum and migration policy, as it was in its infancy as an EU area of harmonization. The EU was creating its structures in a changed environment, in which it was becoming a regional power.  In 2022, the EU has made a concerted effort to demonstrate its unity, and indeed to project power, on all aspects of the approach not only to the humanitarian crisis in Ukraine, but to Russia’s aggression. In spite of many other displacement crises around the world, the EU is, and has to be, laser focused on Ukraine as an existential matter for Europe’s future. The need for the EU to be united in 2022 goes well beyond issues of forms of protection for displaced persons, or any suggestions there are implied hierarchies of refugees. Russia’s aggression towards Ukraine risks destabilizing Europe and current regional security structures in their entirety. The EU must support displaced Ukrainians, it has no alternative.

A second, related, point, which the two situations on which I am reflecting have in common, but which is (ethically) contested, is geography. The Temporary Protection Directive is a ‘first responder’ measure. Although none of the affected former Yugoslav republics directly bordered on a contemporary EU Member State in the 1990s, EU states were immediate neighbours to Yugoslavia. Ukraine is bordered by four current EU Member States. The EU, is by necessity, the location of first response – or one could say, as the EU might describe other continents: the EU is the region of origin. There have been other situations, most prominent among them Europe’s ‘Migration Crisis’ of 2015-16, in which there were calls for the implementation of the Temporary Protection Directive.

The question of ‘why activate the TP Directive now, but not in 2015’ has been posed by many and responses often include a sense that racism or discrimination is at play. That could be the case, or at least part of the answer, whether directly for some member states, or with fears of a backlash from increasingly present right-wing political parties and their followers. However, I would suggest that a major reason for which the directive was not applied in 2015 was that the situation was not an immediate displacement crisis requiring the urgent protection response of direct neighbours. There is a political logic to differing reactions to the immediate displacement of people from a neighbouring state and to the secondary migration of people who, in many cases, were displaced over the course of five years but received limited protection or solution opportunities in other, geographically closer countries (even if a country of first asylum, Turkey, borders the EU). While there are legal and moral obligations to all refugees, there is, of course, a cold hard element of ‘who else could possibly protect them?’ or ‘where else could they possibly go?’ As such, activation could be a result of urgent political decision making in the face of an immediate, proximate crisis in which the EU itself, and most of its Member States as NATO members, has a direct and existential stake. Geography necessarily underpins many decisions in war as in life.

Temporary Protection 1992: Limits on Early former Yugoslav arrivals

In the early 1990s, when former Yugoslav ‘refugees’ started to arrive in European Community (at the time) member states, they faced states focused on erecting barriers to asylum seeker arrivals. European governments had started to realise that in the absence of broad immigration admission channels, the one legitimate way for a person to arrive and remain was to request asylum. In spite of, or even because of, the conflict between Serbia and Croatia, and emerging conflict in Bosnia, those people arriving in various member states were not viewed as refugees in the Convention sense. The focus was on carefully defining, Convention refugee status as a scarce resource, and these people, it was argued, did not fit the individualised refugee definition based on persecution, not war. During the Cold War the people western European nations focused on as refugees were those fleeing Communist countries. Their arrival was limited by the restrictions on exit in the states which they might try to leave. Now they had become the most likely irregular migrants whose arrival, en masse, was feared by governments across western Europe.

In that context, temporary protection was developed as a form of limitation. It offered, by design, a lesser status than asylum: fewer rights, limited duration of stay. It was intended almost as a deterrence, a statement that these people were not refugees, and not expected to stay, and would at best be tolerated. The statuses granted, in terms of rights, residence and reception, differed significantly from one European country to another. Temporary protection in this context essentially signalled ‘you’re here, we can’t send you back, but you’re not welcome’.

Temporary Protection mid-1990s: a provisional, grudging acceptance of Bosnians

Temporary protection evolved somewhat during the 1990s. For some European governments it became the standard. Whereas Convention status was like the ‘gold medal’, preserved for a few, limited and very carefully defined cases, temporary protection was a kind of participation certificate. It manifested as an almost grudging acceptance: towards Bosnians in particular, it was a statement of “well, if you make the journey here, we might have to let you stay because we cannot send you back, but you should really have stayed there or closer to home” (mostly in ‘safe areas’ which proved anything but, or in immediately neighbouring and not yet EU states).

There were continued limitations on rights, reception and residence duration in most European countries, which still took divergent approaches, and still focused on limitations and restrictions. Temporary protection sometimes became a form of provisional protection, a kind of stepping stone to make sure that there really was a longer lasting protection need, as was the case, for example, in the Netherlands.

Temporary Protection 1999: an emergency stop gap for Kosovars

The situation surrounding displacement of Kosovars in 1999 was fundamentally different than it had been for Bosnians in the earlier 1990s, even if the main driving actor (Slobodan Milosevic) and cause (Serb nationalism) were common factors. The Serb forces had started to forcibly displace Kosovar Albanians in Spring 1999, and to kill many. NATO forces intervened to stop this atrocity viewed as undermining the security of the NATO region, even if there were no direct attack on any NATO member state. Hundreds of thousands of Kosovars continued to flee the Serb army’s continued attacks, crossing the borders into Albania and (now Northern) Macedonia. The Skopje government requested assistance as its own ability to shelter the number of arrivals faltered, and it feared a shift in the ethnic balance within Macedonia, which had narrowly avoided its own conflict during the 1990s.

The resulting assistance took the form of a Humanitarian Evacuation Programme, with over 50,000 Kosovars flown out of Macedonia in a few weeks in April/May 1999. These evacuees were, as a group, granted national forms of temporary protection in EU states (some were resettled as refugees to the US and Canada). This could be characterized as an emergency stop gap, which covered not only evacuees but also those who arrived on their own. The lessons of HEP were primarily about solidarity (with a neighbouring state, as well as across the EU) and managing the arrivals of people in need of protection, which contributed to thinking on re-establishing resettlement programmes in the EU.

The absence of significant returns by Bosnians was part of the reason for which there was a lack of generosity in Member States’ approaches to Kosovars who arrived spontaneously, although many had further developed temporary protection policies in the intervening years. The fact was that temporary protection was still meant as a deterrent and restriction. It also meant a focus on return.

There was popular support to do something for Kosovar refugees – NATO countries, overlapping with many EU Member States, had intervened in the conflict. Hostility towards asylum seekers was suddenly, and temporarily reversed, for this case only. Seeing images of people geographically nearby, to whom they could relate, EU populations generally, and briefly, opened their hearts. Governments remained reluctant, however, seemingly aware that the impact of the imagery might not last long, and that a kind of social selfishness would prevail over the longer-term.

Links to and movements towards European Union Asylum Policy

A common feature of all of the Balkan cases is that there was limited EU harmonization or unity. The 1992 Maastricht Treaty on European Union, which initiated ‘harmonization’, was really more a ‘getting to know you’ phase and coincided with the earlier Balkan examples of the use of temporary protection.  In 1999 (Kosovo), the 1997 Amsterdam Treaty was just entering into force and the Finnish Presidency meeting in Tampere setting out the first work programme for the Common European Asylum System was in reality the first move towards more concrete harmonization.

The realization that a European approach to temporary protection was needed came at the conjunction in time of the Tampere work programme to develop the CEAS and the major case of Kosovo, focusing attention of leaders on how working together on managing protection would be to the advantage of most, if not all. The tension was always on the methods of sharing responsibility, be it in terms of relocation or of financial support. 

The 2001 Temporary Protection Directive

The Temporary Protection Directive was the first piece of EU legislation on asylum to be agreed. None of the other elements of the CEAS (e.g. on qualification, reception), which have since been reviewed at least once, were in place. There was the Dublin Convention, but not yet a regulation.

In a sense, therefore, by the time of its first implementation in 2022, the directive was somewhat anachronistic. It had been due for revision, and indeed the Commission had proposed in the 2020 New Pact on Migration and Asylum to repeal the directive and proposed a new regulation to address crisis situations. That initiative was in many ways reactive to events of 2015. However, it is possible that it proves optimal for the EU, the Member States and Ukrainians that the Directive had remained untouched. A primary hypothesis would be that if revised, the provisions of the Directive would have been set more in line with other EU directives in 2022, focused still on an unknown crisis that might require such a measure, and with 2015’s ‘Migrant Crisis’ the most recent major example – one which the EU did not choose to address with the existing temporary protection directive, itself a reflection on the last time there was a displacement crisis from a neighbouring country.

Temporary protection as set out in the directive was no longer about restricting access to asylum. Rather, it was about upholding the integrity of asylum systems in the case of mass influx, while accepting admissions as inevitable due to the nature and location of a major ‘refugee’ producing situation. Indeed, in order to make temporary protection under the directive at least as attractive for individuals as applying for asylum, which of course they have the right to do, the entitlements and rights granted to beneficiaries are closer to those of people with refugee or subsidiary protection status than to those of asylum seekers. Beneficiaries of temporary protection under the directive, if and when activated, can work, study, reside in any EU country.

Temporary Protection 2022: The Welcoming Approach to Ukrainians

Following activation of the Temporary Protection Directive for Ukrainians on 4 March 2022, EU Member States have been able to tell people arriving that they are welcome, that they have access to several rights, including legal residence, healthcare, accommodation, education and employment opportunities, and that they will have this protection for at least one year. In practice, in the initial couple of weeks, the decision to activate the directive has meant the focus can be on the refugees, while behind the scenes the European Commission and Member States prepare operational guidelines and ready systems for registration and the necessary administrative and legal actions associated with residence and protection.

Apart from the actual content of protection in the directive or driven by governments, people in all EU Member States have been offering rooms and housing to arriving Ukrainians. The situation is not without problems: the sheer number of, and chaos of, arrivals has left openings for smugglers and traffickers to prey on people, for example. This is particularly the case as those arriving are primarily women, children and seniors, as able men aged 18-60 are required by the Ukrainian authorities to remain and defend the country. The Directive applies to certain non-citizens also present in Ukraine at the time of the Russian invasion and in need of protection, reacting to some initial instances of apparent discrimination.

The decision to apply the directive grants residence and other rights for an initial year (to 4 March 2023) extendable by two consecutive periods of six months and another year, to make three years in total. National variations on the minimum measures enacted by the decision can take place. Some of these, linked to existing national legislation, might make the actual form that temporary protection takes in some states closer to the temporary protection of the past. But the directive itself, as an EU wide action and all that entails, makes this reactive protection package more proactive and collective than it has ever previously been. The Commission has called for solidarity and transparency, and is pulling together the means to make implementation feasible and well-organized. Time will tell how successful this action is.


On 4 March 2022, a quite exceptional, revolutionary and original way for Europe to approach arrivals from a major displacement crisis in a neighbouring state was activated. Thanks to the Temporary Protection Directive, agreed as a first step in a Common European Asylum System, one that remained hypothetical for over two decades, the EU was ready to act with unity. That alone is a stark contrast with any temporary protection policy? implemented in Europe in the 1990s.

Temporary protection, in the shape of the directive, is an exceptional measure, another major contrast with the way in which the concept of temporary protection was handled in the 1990s. The experience implementing the directive now should lead to modifications that maintain the EU’s readiness to react to any major situation in which European states are necessarily first responders, and continue to enhance the nature of temporary protection as a policy tool for specific situations leading to more entrenched status and protection, rather than as an alternative to Convention or subsidiary refugee protection.

Implementation of the directive will surely not be without problems. Nor will it, as a policy tool, resolve all the problems that could arise in offering protection to those displaced by the invasion of Ukraine. We can hypothesise that when the time comes, the EU might reflect on its response to this displacement crisis, relating it to both the 1990s crises and reactions in 2015, and surmise that one ‘crisis’ instrument will not fit all possible emergencies or crisis models. Whatever the resulting and revised emergency instrument(s), urgent reaction plans, including temporary protection, are clearly a necessary tool among refugee policy instruments that enable states to maintain stability and solidarity in their response, while ensuring access to protection and status for all who need them.