VoxEU Column Migration

Refugees from North Africa: A case for cooperation?

The recent surge of refugees entering Europe from North Africa – particularly through Lampedusa – has caused a rift among EU states over what should be done and who should shoulder the responsibility. This column asks what theory and recent history can tell us about sharing the refugee burden in the EU.

The turmoil in North Africa has led to a surge of migrants across the Mediterranean seeking refuge from the conflict. The relatively peaceful overthrow of president Zine el-Abidine Ben Ali in Tunisia initially led to an influx of 6,000 boat people to Lampedusa, a small Mediterranean island that has served as a holding camp for asylum seekers trying to gain admission to Italy.

As the numbers mounted, the Italian government has sought EU assistance on three fronts.

  • First, it asked for upwards of €100 million to help stabilise the situation on Lampedusa.

  • Second, it urged Frontex, the EU border agency, to strengthen its surveillance of the North African Coast to avert the surge of boat people.

  • Third, it asked for other countries to share the burden through an orderly redistribution of asylum seekers throughout the EU.  

Amid warnings of an influx of over a quarter of a million asylum seekers, Italy was joined by the interior ministers of France, Spain, Malta and Cyprus in calling for assistance from the EU.

In response the EU brought forward Joint Operation Hermes, an Italian-led Frontex operation to strengthen sea patrols in order to detect and prevent illegal landings on the coasts and outlying islands.

On 11 April, Italian Immigration Minister Maroni, invoking a 2001 EU Directive, called for other countries to share the burden by granting temporary protection to some of the refugees. This request was turned down by other countries and the European Commission as premature.

As the number of Tunisians arriving topped 28,000, the Italian government issued them with temporary resident permits. On 17th April France temporarily blocked a train carrying some of the Tunisians over the border from Italy, and it subsequently threatened to suspend its Schengen obligation to allow free movement of those with valid papers (see Boeri 2011 also on this site).

With escalating civil war in Libya and political uncertainty elsewhere in North Africa, is the EU prepared for an escalating refugee crisis on its southern shores?

Haven’t we been here before?

There is an obvious parallel.1 Twelve years ago there was another refugee crisis on Europe’s doorstep—that time in Kosovo. Was the situation then really similar to the present? And if so what lessons can be learned from it? Following the NATO air strikes, tens of thousands of refugees gathered at the Blace border crossing into Macedonia at the end of March 1999. At first Macedonia refused them entry but after a brief hiatus the international community led by the US and Norway under the banner of NATO, and with the aid of the UNHCR and other NGOs, launched a Humanitarian Evacuation Programme. Some refugees were transferred to camps in Macedonia and Albania, and nearly 96,000 were distributed among 28 other receiving countries (an even larger number fled independently). The Humanitarian Evacuation Programme refugees were given temporary refugee status and most returned within the year (these were mainly Albanians and Bosnians; Serbs took longer to repatriate).

These events led to another round of EU soul-searching about developing mechanisms to cope with a sudden influx of displaced persons. The debate about so-called burden sharing goes back to the refugee surge into Germany and Austria that followed the collapse of the Soviet Union. But in the meantime the Amsterdam Treaty had begun to shift the locus of decisions over asylum and immigration towards the EU and away from national governments. One outcome was the Temporary Protection Directive of 2001 (noted above), which laid down standards and procedures for the reception of those given temporary protection. In the event of a sudden influx it called for a “balance of effort” among member states in a spirit of solidarity, but it provided no burden sharing formula or mechanism.

Isn’t it just politics?

Is there any basis for refugee burden-sharing among host countries other than some notion of fairness in distributing unwanted refugees more evenly when it is not possible to shut them out completely? In fact there is a good deal of popular support for providing a safe haven for genuine refugees fleeing individual persecution or generalised violence. Satisfying altruistic motives towards refugees by providing them with sanctuary is a form of benefit to the population of host society. And individuals in one country might be expected to derive some utility from the knowledge that refugees find safety in other countries. As this benefit is non-rival and non-excludable, hosting refugees is (to some degree) providing a public good. But there are also financial and social costs that fall exclusively on the country that provides asylum.

A simple model that captures these notions assumes that there is diminishing marginal utility from refugees and a constant cost per refugee. When countries set their policies independently they fail to take account of the benefit of their refugees to the citizens of other countries. Thus in the Nash equilibrium refugee places will be underprovided. In each country policy will be tougher than if policy in all countries was decided by a social planner. And the more uneven is the pressure of asylum seekers across different countries the greater is the shortfall. The willingness to accept refugees might be increased by providing a financial subsidy to reduce their marginal cost. But a severe imbalance could only be addressed by some form of direct redistribution of refugees among the host countries.

Policies to redistribute refugees would need to be implemented by a centralised authority. But would the citizens of the EU be prepared to cede the authority for asylum policies more fully to what some see as faceless bureaucrats in Brussels? In 2002, after the last great surge of asylum applications, the European Social Survey asked respondents whether they would prefer immigration and asylum policies to be made at the national/local level, or at the international/regional level. Fifty-seven-and a-half percent preferred that policy should be made at the supra-national level, so opposition to centralised policymaking is not as great as is sometimes believed. Anti-immigration pressure often comes from extreme right wing parties and is whipped up by the press so that mainstream parties have to respond by talking tough and acting tough. Hence there is a potential political benefit, at least to the more moderate citizens and the parties that represent them, of taking asylum out of national politics and shifting more of the responsibility to the EU level.

Signs of progress?

The refugee crisis in North Africa is different from the Kosovo crisis in several ways. This time the US has taken more of a back seat in the NATO operations and it is much less likely to take the lead in evacuating refugees or in brokering a deal with other countries for their temporary protection. At the time of Kosovo the fledgling state of Macedonia (not an EU member) was far less able to cope with and influx of refugees than are countries such as Italy and France (perhaps not Malta). This time the responsibility to act falls more squarely on the EU.

In fact, the EU is much better organised to respond than it was before the Kosovo crisis. In 2000 the European Refugee Fund was set up to provide subsidies to individual countries for refugee care. Since its founding the European Refugee Fund has been beefed up and it can be used to provide financial support in refugee emergencies. The most recent step is the setting up of a European Asylum Support Office which is located in Malta and began operations at the end of 2010. This office is tasked with fostering the exchange of information and the dissemination of best practice methods as well as establishing an early warning system and mechanisms for assisting states that are under “particular pressure. Interestingly it is also expected to assist in the relocation of recognised refugees, but only on an “agreed basis” between member states and with the consent of the individuals concerned.

What is missing is any formula or mechanism for distributing refugees across the countries of the EU. In the light of recent events, and consistent with the theory outlined above, it seems unlikely that individual countries that are not directly affected will volunteer to accept significant numbers. The powers of the European Asylum Support Office need to be expanded so that it can lead rather than follow. And it needs to be done sooner rather than later.


Boeri, Tito (2011), “Reforming Schengen is the wrong response to the Lampedusa crisis”, VoxEU.org, 29 April.

1 I am currently preparing a report on “Seeking Asylum in the OECD” as part of the CEPR’s research programme on Politics, Economics and Global Governance: The European Dimensions (PEGGED). 


210 Reads