Who is a refugee? Regulatory approaches.

When regulating the domain of asylum, it is advisable to build on existing international criteria or to make use of specific regional instruments. We examine these in the first part of this article (I.) Furthermore, we present here some national regulations with different degrees of complexity that can serve as reference (II.). Moreover, we highlight some particular provisions that might be of interest to regulators (III.). Finally, we analyse in more detail the proposal of the European Commission for a “Qualification Regulation” (IV.).

  1. International and regional instruments

1. The Geneva Convention (1951) and its Protocol (1967) relating to the Status of Refugees should serve for its Signatories as the minimum basis for further regulation at national level. The most relevant provision can be found in the Article 33 of the Convention, which imposes on the states the prohibition of non-refoulement. The Convention therefore protects a refugee from being expelled or returned to a territory ” where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion.”

The Convention also contains the definition of a refugee, nevertheless, it does not contain more detailed provisions regarding the reasons for persecution or the acts and actors of persecution. Therefore, the UNHCR Handbook could serve as a further orientation document.

2. Regulators could also get inspiration from the OAU Convention governing refugee problems in Africa for which “external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of his country of origin or nationality” amounts to a refugee recognition criteria.

3. Additionally, the Cartagena Declaration on refugees (Central America, Mexico) recognises a threat caused by “generalized violence, foreign aggression, internal conflicts, massive violation of human rights…” as reason for attributing protection.

4.a) At EU level, the Qualification Directive contains even more detailed provisions both on the acts of persecution (Article 9), but also the reasons for persecution (Article 10). It contains definitions of the concept of race, religion, nationality, social group and political opinion.

b) The European Commission has recently proposed a new act, amending and converting it into a directly applicable Qualification Regulation. It contains relevant definitions and detailed provisions on qualification criteria for both refugees and beneficiaries of subsidiary protection. The rest of the so-called Qualification Regulation is devoted to revocation/review of the protection status and content of protection rights and obligations.

The definitions from the Directive remain valid whilst some new provisions can be found in the Regulation. Some provisions have become more lenient. It is for example no longer expected from an applicant to behave discreetly or abstain from practices inherent to his/her identity in order to avoid persecution (Article 10(3) of the proposed Regulation). Conversely, the EU member states shall now waive the non-refoulement prohibition when there is an “internal flight alternative” (see the shift from “may” to “shall”). Regarding the review and revocation of the refugee status, the EU member states now have an obligation to phase-out the protection when the conditions in the Regulation for revocation are met; there is no room for the application of the proportionality principle anymore. In this context, the Regulation introduced a rigid definition of “serious non-political crimes” (Article 12(5) of the draft Regulation). However, there is no definition of “particularly serious crime” and no provision on when exactly a refugee can be qualified as a “danger to the security of the Member State”. This is likely to lead to diverging interpretations and thus to an unharmonised administrative practice.

  1. National reference regulation

In the following section, we will present different types of national legal frameworks on refugees in view of diverse quality criteria of a regulation, namely its completeness, its complexity, its coherent level of depth, its structure, its length and relevance as possible reference for other jurisdictions.

All legal frameworks examined hereafter contain broad range of provisions on refugees. Thus the regulations do not only deal with what is in our focus (the recognition criteria), but also with the rights given to refugees, the reception conditions and on the asylum procedure itself. However, we will limit our analysis on the protection criteria and other pertinent provisions that are closely linked to this domain.

For the purpose of finding a suitable regulatory framework, we investigated into several legal systems aiming to identify “light”, “medium”, “upper-medium” and “complex” regulation.

The international benchmarks are already at “medium” level with the Proposal for the EU Qualification Regulation being “upper-medium”. Likewise, the majority of national regulations examined can be classified under these two categories.

The vast majority of legal systems directly incorporates refugee recognition criteria from the Geneva Convention. Therefore we could note many similarities. Nevertheless, there are still countries with no legal framework in place (for ex. Maldives, Malaysia) and countries that only grant political asylum (for ex. Botswana, Lebanon). We believe that these regulations are too limited in scope to serve as a regulatory models.

The rest of the national legal frameworks examined differ on the range of topics covered, the level of details, its structure and complexity. Some states have specific regulations on political asylum and refugees. Others integrate the refugee protection criteria in their national regulations on aliens or immigration, which often include a wide range of provisions, from visas to human trafficking. Furthermore, in many national systems the protection granted by the Geneva Convention is complemented by other categories of protection, such as subsidiary protection, humanitarian protection, protection “sur place”. Thirdly, some national regulations contain cessation and exclusion clauses, i.e. “negative” criteria excluding a person from refugee protection. Such exclusion clauses are laid down in Article 1E, 1D and 1F of the Geneva Convention, but some states extend those provisions to further cases.

1. The regulation of New Zealand is a good example for a “light” regulation. Its eligibility criteria for refugees are not very detailed. The regulation mainly refers to the definition laid down in the Geneva Convention. Such a regulation requires the reader to consult the international text in addition to the national regulation which is not very “reader-friendly”. Yet, New Zealand provides two additional types of international protection by referring to the obligations under the Convention Against Torture and the 1966 International Covenant on Civil and Political Rights. The authorities are therefore not allowed to deport a person, who would be in danger of being subjected to arbitrary deprivation of life, torture or cruel treatment. Beyond the field of qualification criteria, e.g. asylum and immigration procedures, visas and penalties, the New Zealand regulation could be classified in the category of “upper-medium” complexity. There are very detailed provisions on these issues.

2. The regulation of Venezuela can be classified in the category of “lower-medium” regulation. It replicates the content of the Geneva Convention, while also granting proper political asylum. The regulation of Venezuela can be commended for being easily comprehensible law. However, it does not contain detailed provisions. E.g. it does not contain provisions on the grounds of persecution in the country of origin.

On the same level of detail and complexity, we could equally commend the law of Burundi, as it could be a good reference for those who would wish to regulate on the refugee recognition criteria in French. In addition, Burundi has set up a provision offering protection to a failed asylum-seeker in case of serious personal distress. Moreover, the asylum seeker’s family situation and her/his integration in the society are taken into account.

3. South Africa has a “medium” complex asylum law, with very clear provisions on the refugee recognition criteria. It is fairly similar to the international benchmarks, while imposing the prohibition of non-refoulement even in cases where a person’s life/safety/freedom would be threatened due to external aggression, occupation, foreign domination or other events seriously disturbing or disrupting public order.

4. Portugal can be classified in the “upper-medium” category and has of one of the most complete, but not overly complex laws on the refugee recognition criteria, as it defines broad range of reasons for persecution. It also enlists acts and actors of persecution, and contains cessation and exclusion clauses. Portugal explicitly foresees judicial review in case a cessation clause would become applicable for a recognised refugee. In addition, it confers special attention to victims of torture or rape. Even if we look at the Portuguese law more broadly, it regulates the resettlement of refugees, contains elaborate provisions regarding the content of the refugee status (for ex. access to housing), and extends effects of the asylum through the family unity principle to dependent persons of a refugee. Whilst being so complete, the Portuguese regulation is neither excessively detailed, nor too complex nor too lengthy.

5. Australia has a very complete law on the refugee protection criteria, including detailed provisions on the determination of a refugee, well-founded fear of persecution, membership of a particular social group, and particularly serious crime. However, we could note that the Australian law is fairly restrictive, as it grants protection only to third-country national whose “real chance of persecution relates to all areas of the receiving country”, if they would be sent back. This reflects the practice of many EU Member States which do not recognise asylum seekers who could have fled to “safe” areas in their country of origin.

6. Canada has adopted an exhaustive law on the protection criteria. It grants protection both under the Geneva Convention and the under the Convention against Torture. In addition, a person who would be subjected to a risk to their life/cruel and unusual treatment/punishment in a third country is offered protection. Furthermore, a member of a “class of persons” falling under the criteria can also benefit from the protection. Canada also thoroughly regulates the exceptions to the prohibition of non-refoulement, e.g. by defining a “serious criminality”. Speaking more generally, the Canadian law contains very precise provisions regarding the rules of entry to the territory and the competent authority. It also devotes a lot of attention to the enforcement, and can be classified as a well-structured model of “complex” legislation. Moreover, Canada envisages that more detailed regulations shall address certain aspects of the prohibition of non-refoulement, namely by creating and updating a list of safe third countries, that respect both the Geneva Convention and the Convention Against Torture.

  1. Particular provisions

Gambia contains a clause on prima facie refugee recognition criteria:

“…the Secretary of State may, …, declare a person or persons belonging to a particular class or nationality, on the basis of objective, prevailing circumstances in that person’s or those persons’ country of origin or nationality, to be a refugee or refugees recognised on a prima facie or group basis.”

The principle of prima facie or group recognition is used in many countries in Africa to avoid cumbersome recognition procedures where there is a very high likelihood of political persecution, e.g. against certain tribes or religions. However, it is also from time to time used by rich countries to manage a strong influx. Such was the case during the Bosnian civil war or at the time when Kosovars were persecuted by Serbia (when the Kosovo was still part of Yugoslavia). Prima facie recognition is thus a useful tool to reserve scarce administrative capacities for countries of origin for which it is more complex to assess whether applicants have a well-founded fear of political persecution.

Based on this idea, Ireland recognises a special category of so called “programme refugees”. This type of protection is given to a person who is allowed to remain in the state for temporary protection by the permission of the Government. The regulation of Moldavia contains even more elaborated provisions on prima facie recognition, such as guarantees to unaccompanied minors, procedural aspects and has a well-formulated cessation clause (“with due observation of human rights, fundamental freedoms and non-refoulement obligations”).

Burundi and South Africa recognise as refugees persons who are dependent on the applicant, namely the spouse, the unmarried dependent child or any destitute. Generally, states only grant them residence permits or extend effects of asylum through the family unity principle.

Sweden has adopted a provision protecting victims of environmental disaster.

Georgia, on the contrary, excludes environmental or economic reasons from the refugee recognition criteria.

For Switzerland an alien who has a well-founded fear of persecution due to his refusal to perform military service in his country of origin is not a refugee, whereas judges in some EU member states recognise asylum seekers who refuse the military service for reasons of political or religious consciousness.

Furthermore, Russia and Kazakhstan laws contain a provision, which excludes from the refugee protection certain applicants (below Kazakhstan provision):

5) there are reasonable grounds to believe the applicant took part in the activities of terrorist, extremist or banned religious organisations functioning in the country of origin or country from which he/she has arrived;

Azerbaijan and Moldavia have set-up a provision calling for international cooperation with other states and international organisations with the aim to find solutions to the problems in the field of asylum.

Finally, one could look at Canada’s Regulations, which could serve as a reference for further and more detailed formulation of regulatory acts on the refugees (for ex. to comprehensively address the data protection of a refugee). Evidently, these regulations go largely beyond the topic of this article, the qualification criteria.

  1. Comments on the Proposal for the EU Qualification Regulation

As already noted, the European Commission Proposal for a Qualification Regulation can be classified among “upper-medium” regulations. Without taking a position on the substantial aspects of the proposed Regulation, we believe that it has a coherent level of depth and is generally of a high technical quality.

The proposed Regulation clearly stipulates its objective to harmonise rules on asylum at EU level, namely by introducing common recognition criteria and uniform status and rights across the EU. It thereby complements the EU “Asylum Procedure Regulation” and “Reception Conditions Directive” or even creates a basis for these two acts. Whereas most national legal frameworks cover all the three domains in one act, the EU regulates them separately, which is not necessarily negative as the domain is broad and complex. However, when analysing the completeness of the Qualification Regulation, one needs to consult the other two documents. E.g. we thought initially that provisions on the cooperation among states or on the data protection were missing in the Qualification Regulation until we found these provisions in the other two acts.

How could the proposed Qualification Regulation be complemented or otherwise improved?

First, the Qualification Regulation does not contain any provision on prima facie or group refugee recognition. The absence of such provisions can at first sight be explained by the fact that the EU disposes of a separate Temporary Protection Directive. However, this Directive is quite vague and can only become applicable after the Council of the EU adopts a decision by a qualified majority. In the past years, we could observe obvious “mass influx” situations inside the EU, when the asylum claims exceeded the capacity of national asylum systems. Nevertheless, the Council has never decided to trigger the application of the 2001 Directive. As the Temporary Protection Directive does not seem to work as expected, there is even more room for considering provisions on prima facie or group refugee recognition than there would be otherwise.

But even if this Directive was to be functional, there would be room for complementing the Qualification Regulation because the Temporary Protection Directive does not apply in situations other than those of mass influx. E.g., it is not applicable where states or state-like structures persecute systematically certain ethnic or religious minorities without this triggering a mass influx in the EU. In these situations of doubtless prima facie or “group persecution”, it does not make sense to re-traumatise the refugee by inquiries on the personal persecution history and to waste the scarce administrative resources of the asylum administrations for investigating and assessing the individual asylum claim. Instead, the individual should already be recognised because s/he belongs to the persecuted group. The only element to verify in the asylum procedure is the question whether the asylum seeker belongs indeed to that group or not. Once it is assured that the asylum seeker belongs to the group, s/he should be singled out from the ordinary cumbersome procedure and take profit from immediate recognition or at least from a kind of procedural facilitation. In order to implement such immediate recognition of procedural facilitation correctly, it is useful if not even necessary to establish a separate criterion “group persecution”. We consider that such criterion could best be added to the Qualification Regulation. The other two acts proposed by the European Commission could then link to this criterion.

We believe that the EU Regulation could in addition mirror the law of South Africa, by offering protection to persons facing external aggression, occupation, foreign domination or events seriously disturbing public order in either a part or the whole third country. In this context, we would like to commend reading the OUA Convention and the Cartagena Declaration already mentioned in Section I. of this article.

Following the model of Sweden, the Qualification Regulation could also contain a refugee recognition criteria due to environmental disasters in his/her country of origin. This type of provision is certainly very pertinent for the ever growing world displacement as a result of the climate change.

One crucial topic of law-suits in the field of political asylum is the question whether actors others than states may be the authors of persecution qualifying asylum seekers as refugees. The Qualification Regulation, by its Articles 6 and 7, is very generous in this sense and even recognises persecution by private actors provided that the state or the “party” or “organisation controlling the state or a substantial part of the territory of the State” are either unable or unwilling to provide protection. It is to be admitted that this is a very much appropriate and fine-tuned approach. The only minimal pitfall is that some situations might not be appropriately reflected by this wording. There are state-like structures in the world which sometimes persist for many years, see Abchasia and Somaliland. But are these “organisations”? Agents of the asylum authorities might have diverging views. Hence the formulation merits improvements so as to create a continuum going from (private) organisation to full-fledged state, leaving no case in the loophole in between. The following cases might be used to check whether an alternative wording would really appropriately cover all situations:

  • During civil wars, there can be different state-like power structures which neither qualify as “state” nor as “organisation”, see “Puntland” in North-Eastern Somalia today or the UNITA hold area in Angola during its long-lasting civil war.

  • Likewise, there are countries in which a certain regional state-like structure holds state powers on a certain territory whereas the central state holds little or virtually none, see e.g. the Kurdish region in Iraq today.

  • In some parts of the world, organised crime is not only able to steer the administrations of the state in such a way that criminals have impunity and their victims cannot find protection, but holds even state-like structures on its own, including penal courts and tax administration. Likewise, some terrorist organisations hold state-like powers on a certain territory. Are these organised crime or terrorist organisations meant by Article 7? Supposedly not, but this merits clarification.

Moreover, we commend introducing a provision on when exactly a refugee can be qualified as a “danger to the security of the Member State”. The absence of these two definitions is likely to lead to diverging interpretations and thus to an unharmonised administrative practice.

The Qualification Regulation contains an Article 8 which excludes recognition in cases where the asylum seeker could find so-called “internal” protection in a certain area of his country of origin. Whilst the concept is certainly justified in principle, there are details to be taken into account which make the application of this concept difficult. The proposed Qualification Regulation covers admittedly many aspects to be taken into account. However, it misses the question which is the most relevant one in practice: Which minimum level of subsistence must be ensured in the safe part of the country of origin so that the asylum seeker can really be asked to search for protection there? The Qualification Regulation only requests to examine whether a stay in the safe part of the country of origin would “impose undue hardship”. This is pretty vague an expression when bearing in mind that the vast majority of countries of origin have no social system ensuring economic survival. Is a life in the desert, protected only by a UNHCR tent and insufficient food supply an “undue hardship”? Is it already an “undue hardship” to start from zero without housing and job in another city marked with a high unemployment rate for those having the profession of the applicant and a high general unemployment rate? To avoid a diverging decision practice of EU member states, more specific provisions are needed.

Evidently, the same considerations are also applicable to the exclusion clause “safety in a third country” which is inter alia referred to in Article 12(1) of the proposed Qualification Regulation. Even when an applicant is, in a third country, protected against refoulement to his country of origin, that does not mean that s/he can make a living there. The many Syrian refugees in Jordan who have no or too limited access to the labour market and sometime even feel compelled to sell their daughters into prostitution or premature marriage, exposing these to exploitation and domestic abuse, can testify of this.

Finally, we miss provisions on the requested minimum likelihood of persecution or harm. Based on the Geneva Convention, the Qualification Regulation requests a “well-founded fear” or a “real risk”. But starting from which degree of likelihood a certain fear is “well-founded” and a risk becomes “real”? Is a 10% likelihood sufficient? Shall 50% be required? Shall different degrees of likelihood be applied to different acts of persecution or risks? Not regulating on this question will leave an extremely broad range of interpretation open to each individual agent of the asylum authorities and to the judges reviewing the authorities’ decisions. It is not to be excluded that the main reason for the currently very much diverging decision practice of asylum authorities in the EU and even sometimes within the same EU member state is due to the fact that the agents and judges require different degrees of likelihood.

Further links

If you wish to investigate further national regulations on asylum, please check the following website of the United Nations High Commissioner for Refugees:


This article has been written by Ajda Mihelčič, M.A.S., on behalf of the Regulatory Institute, Brussels.

Leave a Reply

Your email address will not be published. Required fields are marked *

5 × four =