In Gudanaviciene & Ors, R (on the application of) v The Director of Legal Aid Casework & Ors [2014] EWCA Civ 1622, the Court of Appeal considered the Lord Chancellor's appeal against the High Court's earlier decision that the exceptional funding guidance was unlawful, and that refugee family reunion was in scope of Schedule 1 of LASPO. The Court of Appeal refused to follow the approach of the High Court either in Gudanaviciene or the earlier cases of M v Director of Legal Aid Casework [2014] in construing s10 of LASPO (the section dealing with exceptional cases. Lord Dyson MR, giving the judgment of the Court, set out the test to be applied thus: We see no warrant for construing section 10(3)(a) as imposing a condition that an ECF determination should only be made where it can definitely be said (Coulson J's formulation) that refusal would be a breach; or where there is a "high level ofprobability" that refusal would be a breach (Collins J's test). There is no need to add a gloss to the wording of the statute "would be a breach". In deciding whether there would be a breach, the Director should apply the principles to be derived from the case-law (some of which is mentioned at para 27 of the Guidance). There is no need for elaboration. When determining whether a complaint of a breach of Convention rights has been established, the ECtHR does not ask itself whether there has definitely been a breach or whether there has been a breach to a high level of probability. It simply asks whether there has been a breach. In our view, this approach should inform the meaning of the words "would be a breach" in section 10(3)(a). We do not consider that the word "clearly" in the Explanatory Notes (see para 9 above) takes the argument any further. We should add that we accept the submission of Mr Chamberlain that the "real risk of a breach" is a concept which has no part to play in the exercise envisaged by section 10(3). Section 10(3)(a) speaks of the situation where a failure to make civil legal services available would be a breach, not where there would be a real risk of a breach. The concept of real risk has no part to play in the question whether the denial of legal aid would amount to a breach of an individual's procedural rights under the Convention or under article 47 of the Charter.

In short, therefore, if the Director concludes that a denial of ECF would be a breach of an individual's Convention or EU rights, he must make an exceptional funding determination. But as we shall see, the application of the ECtHR and CJEU case-law is not hard-edged. It requires an assessment of the likely shape of the proposed litigation and the individual's ability to have effective access to justice in relation to it. The Director may conclude that he cannot decide whether there would be a breach of the individual's Convention or EU rights. In that event, he is not required by section 10(3)(a) to make a determination. He must then go on to consider whether it is appropriate to make a determination under section 10(3)(b). In making that decision, he should have regard to any risk that failure to make a determination would be a breach. These words mean exactly what they say. The greater he assesses the risk to be, the more likely it is that he will consider it to be appropriate to make a determination. That is because, if the risk eventuates, there will be a breach. But the seriousness of the risk is only one of the factors that the Director may take into account in deciding whether it is appropriate to make a determination. He should have regard to all the circumstances of the case. In considering the Lord Chancellor's Guidance in relation to whether there would be a breach of Art 6 ECHR in refusing funding, Lord Dyson said: In our judgment, the cumulative effect of these passages is to misstate the effect of the ECtHR jurisprudence. As we have seen, the Guidance correctly identifies many of the particular factors that should be taken into account in deciding whether to make an exceptional case determination, but their effect is substantially neutralised by the strong steer given in the passages that we have highlighted. These passages send a clear signal to the caseworkers and the Director that the refusal of legal aid will amount to a breach of article 6(1) only in rare and extreme cases. In our judgment, there are no statements in the case-law which support this signal. For the reasons stated earlier, we do not consider that the reference in X v UK to "exceptional circumstances" provides support for it.

The general principles established by the ECtHR are now clear. Inevitably, they are derived from cases in which the question was whether there was a breach of article 6(1) in proceedings which had already taken place. We accept the following summary of the relevant case-law given by Mr Drabble: (i) the Convention guarantees rights that are practical and effective, not theoretical and illusory in relation to the right of access to the courts (Airey para 24, Steel and Morris para 59); (ii) the question is whether the applicant's appearance before the court or tribunal in question without the assistance of a lawyer was effective, in the sense of whether he or she was able to present the case properly and satisfactorily (Airey para 24, McVicar para 48 and Steel and Morris para 59); (iii) it is relevant whether the proceedings taken as a whole were fair (McVicar para 50, P,C and S para 91); (iv) the importance of the appearance of fairness is also relevant: simply because an applicant can struggle through "in the teeth of all the difficulties" does not necessarily mean that the procedure was fair (P,C and S para 91); and (v) equality of arms must be guaranteed to the extent that each side is afforded a reasonable opportunity to present his or her case under conditions that do not place them at a substantial disadvantage vis-à-vis their opponent (Steel and Morris para 62). In making an exceptional funding determination, what is required is a consideration of all the circumstances of the case (and "exceptional" is not a test of itself, nor does it necessarily imply that grants will be rare, contrary to what is stated in the Guidance). The Strasbourg caselaw doesn't require representation in all but the most straightforward of cases; but nor does it only require representation in extreme cases - "the critical question is whether an unrepresented litigant is able to present his case effectively and without obvious unfairness. The answer to this question requires a consideration of all the circumstances of the case, including the factors which are identified at paras 19 to 25 of the Guidance. These factors must be carefully weighed. Thus the greater the complexity of the procedural rules and/or the substantive legal issues, the more important what is at stake and the less able the applicant may be to cope with the stress, demands and complexity of the proceedings, the more likely it is that article 6(1) will require the provision of legal services (subject always to any reasonable merits and means test). The cases demonstrate that article 6(1) does not require civil legal aid in most or even many cases. It all depends on the circumstances." (para 56). The Court went on to consider Art 8 in immigration cases.: Whether legal aid is required will depend on the particular facts and circumstances of each case, including (a) the importance of the issues at stake; (b) the complexity of the procedural, legal and evidential issues; and (c) the ability of the individual to represent himself without legal assistance, having regard to his age and mental capacity. The following features of immigration proceedings are relevant: (i) there are statutory restrictions on the supply of advice and assistance (see section 84 of the Immigration and Asylum Act 1999); (ii) individuals may well have language difficulties; and (iii) the law is complex and rapidly evolving (see, for example, per Jackson LJ in Sapkota v Secretary of State for the Home Department [2012] Imm AR 254 at para 127).

Para 59 is plainly correct: immigration decisions do not involve the determination of civil rights and obligations. But para 60 is wrong as Mr Chamberlain has conceded. For the reasons that we have given, the W v UK test should be applied in immigration proceedings.

What guidance is it appropriate to give as to the circumstances in which article 8 requires the provision of legal aid in immigration cases? We have already set out at para 72 above some of the relevant circumstances. In addressing these, it will often be helpful to take into account the factors set out at paragraphs 19 to 24 of the Guidance in relation to article 6(1). In carrying out this exercise in relation to article 8, the decision-maker should not apply a "very high threshold" for the reasons that we have given in rejecting such a threshold in relation to article 6(1).

Deportation cases are of particular concern. It will often be the case that a decision to deport will engage an individual's article 8 rights. Where this occurs, the individual will usually be able to say that the issues at stake for him are of great importance. This should not be regarded as a trump card which usually leads to the need for legal aid. It is no more than one of the relevant factors to be taken into account. The fact that this factor will almost invariably be present in deportation cases is not, however, a justification for giving it reduced weight. It then considered the various individual appeals, which included family reunion, deportation and trafficking cases. Practitioners dealing with such cases will want to consider those parts of the judgment carefully. The Court did overturn the conclusion of Collins J in the High Court that refugee family reunion was in scope. It found that family reunion is not a right "arising from" the Refugee Convention, and so not covered by para 30 of Schedule 1 of LASPO, and Parliament did not intend otherwise. This means that family reunion cases must now be considered outside the scope of legal aid again. At the time of the High Court judgment, the LAA put a statement on its website (now only available in the National Archives version). It referred back to this statement in a new (if not exactly clear) note following the Court of Appeal case. It confirms that the LAA will pay for work done on cases opened since the High Court judgment - but only up to the date of the Court of Appeal judgment (15 December 2014). Any further work done after that date (regardless of when the case began) will not be paid for unless exceptional funding is granted.