Asylum Aid has published a new report, Even If… The use of the Internal Protection Alternative in asylum decisions in the UK, analysing the reasoning on internal protection or internal relocation in Home Office asylum decisions, identifying a number of failings appearing frequently in the decisions read for the research. To lawyers practising in asylum, the flaws in reasoning will be familiar, but it is useful to have this detailed analysis of patterns emerging from thematic analysis of a number of refusal letters.

Their overview of decision letters suggested that decision makers frequently fail to identify new risks in the purported area of protection, particularly for those who belong to marginalised groups such as women and LGBTI individuals. New risks are treated as matters going to reasonableness instead of being considered as Article 3 risks. This goes along with the use of the term “internal relocation” rather than “internal protection”, the latter relating expressly to the person’s safety.

Weighing up levels of harshness which the applicant can be expected to tolerate is unlawful if there is a real risk of serious harm.

They found that reasoning is often backward-looking (“you did not seek the protection of the state authorities, therefore you cannot say they would not have protected you”) whereas the question of internal protection is inherently forward-looking: will the person be safe in future if they return? In some cases, the “failure” to seek protection from the state authorities was taken against credibility of the fear itself, not merely as a part of the consideration of whether protection would be available in future.

Where a refugee has not sought the protection of the authorities or not moved elsewhere in their own country, there is a gap in the evidence as to what would happen if they now did that. This may be filled by a combination of their own testimony and COI. It cannot be filled by treating internal protection as something that should have been tried, and therefore should now be relied on.

They noted a failure by the Home Office to properly address the issue of how a person might be found by a would-be persecutor in the country of return, pointing out that this information is rarely covered in Country of Origin reports. There was a need for local and anthropological knowledge to be taken into account.

Often the main source of evidence was the applicant themselves. However, in the files analysed, leads given by the applicant as to the reach of the persecutor were not followed up by COI. They were rejected on the basis of ideas about the persecutor or the country of origin which were untested with the applicant and not grounded in the context of the country in question.

Another issue arising was Home Office reliance on largely irrelevant Country of Origin information, such as the existence of NGOs campaigning on the issue in question, descriptions of government bodies, the size and structure of the police force and legal provisions without understanding of “the real power structures” or whether the protection existed in practice. The report notes that

In informal conversation during the research, asylum seekers expressed puzzlement that the Home Office considered their education would keep them safe.

In that respect the analysed decisions showed that:

• The comments of the applicant in interview might not be given any weight;
• The refusal letter applied COI selectively. Even when COI substantiating the applicant’s assertion was quoted, the link was not made, and the COI might even be used to opposite effect;
• Leads given by the applicant were not followed up by COI research.

In particular the research showed that Home Office decisions often conflated the safety limb and the reasonableness limb of the test for an internal protection alternative “and this obscured the possibility that reliance was placed on an NGO as an actor of protection.” Likewise where the Home Office relied on family support as the basis for an IPA, the writers point out that this reasoning is lawful only if the presence of family mean that the persecutor would not target the person. If, on the other hand, the family is expected to protect the person from the persecutor despite targeting, “this would cast the family as an actor of protection, which would not be legitimate under the Refugee Convention or QD Article 7.”

The report’s conclusion highlights the risks inherent in placing the whole burden of proof on the applicant as well as the flaws in Home Office reasoning which often fails to reach the root of the most important question: will this person be safe?