State Protection Under s.31 International Protection Act 2015: “Improvements” Are Not “Effectiveness” — COI Findings Must Rationally Support the Conclusion

1) Introduction

M.G. (Georgia) v The International Protection Appeals Tribunal and Anor [2026] IEHC 151 is a High Court judicial review of a decision of the International Protection Appeals Tribunal (“the Tribunal”) dated 7 August 2024 affirming an International Protection Office (“IPO”) recommendation to refuse the applicant asylum and subsidiary protection.

The applicant, a Georgian national, feared persecution/serious harm arising from sustained domestic violence and coercive control by her ex-partner. Notably, both the IPO and the Tribunal broadly accepted her account, accepted that the harm reached the level of persecution, accepted that her fear was well-founded, and accepted the Convention nexus (membership of a particular social group: victims of domestic violence / women in a long-term relationship with a violent partner). The refusal turned instead on one issue: whether effective state protection would be available in Georgia on return, assessed largely through country-of-origin information (“COI”).

The case therefore sits at the intersection of (i) the statutory state-protection test under s.31 International Protection Act 2015 (“the 2015 Act”), (ii) the Tribunal’s obligation to fairly appraise COI, and (iii) the supervisory (not appellate) role of the High Court in judicial review.

2) Summary of the Judgment

O’Donnell J. quashed the Tribunal’s decision and remitted the appeal to a differently constituted Tribunal member. The applicant did not succeed on all grounds: the Court rejected challenges based on (i) inadequate reasons, (ii) internal inconsistency (as a standalone quashing error), and (iii) material error of fact, though it acknowledged some avoidable confusion in the Tribunal’s narration of post-2015 threats.

The applicant succeeded on the central ground: the Tribunal’s state-protection finding was unlawful because it did not rationally flow from the COI and because the Tribunal conflated “improvements” with “effectiveness”. In the Court’s view, the COI painted, at best, a “mixed picture” with a general tenor of continuing inadequacy in the response to gender-based violence; it did not reasonably support the Tribunal’s conclusion that Georgia provided effective, non-temporary protection meeting s.31.

3) Analysis

A) Precedents Cited

i) Core statutory framework and the “four-part test”

The Tribunal (and the High Court) relied on B.C. v. IPAT [2019] IEHC 763 as summarising how s.31 of the 2015 Act is operationalised. While the High Court did not restate the test verbatim in this judgment, it treated B.C. as anchoring the structured inquiry into whether the state takes reasonable steps, operates an effective legal system, provides protection that is effective and non-temporary, and whether the applicant has access to it.

ii) Reasons and the scope of explanation

On reasons, O’Donnell J. applied established Irish authority that decision-makers need not address every point in narrative form, but must engage with core claims and disclose a comprehensible reasoning pathway:

  • G.K. v IPAT [2022] IEHC 204 (reasons need not be discursive; engagement required with core issues).
  • YY v Minister for Justice and Equality [2017] IESC 61 (reasons must allow the affected person and the court to understand the decision-making path).
  • Middelkamp v Minister for Justice and Equality [2023] IESC 2 (reasons/justification sufficient to permit meaningful review).

Applying these, the Court held the Tribunal’s reasons were adequate—even though the ultimate state-protection conclusion was not lawfully supported by the evidential base.

iii) “Considered all the materials” and the presumption of regularity

The respondents invoked the line of authority stemming from G.K. v Minister for Justice [2002] 2 IR 418, supporting the proposition that where a decision-maker states it considered the materials, the court should not lightly infer the contrary. The Court accepted the general principle but held that it does not immunise a decision where the COI analysis does not rationally sustain the conclusion required by s.31.

iv) COI appraisal, selectivity, and resolving conflicting material

The judgment’s most important doctrinal work is its synthesis and application of prior High Court authority on unlawful COI treatment:

  • D.V.T.S. v. The Minister for Justice [2008] 3 IR 476: Edwards J. quashed a decision where COI was used selectively and conflicting information was not rationally analysed or explained. O’Donnell J. treated D.V.T.S. as establishing that where COI conflicts, the Tribunal must (a) recognise the conflict and (b) justify why one strand is preferred.
  • B.A. v. IPAT [2020] IEHC 589: Burns J. quashed a state-protection finding where the Tribunal’s own COI summary (showing serious legal/practical deficits) could not rationally lead to the positive protection conclusion. O’Donnell J. treated B.A. as illustrating irrationality where “conclusion and COI” do not cohere.
  • F.M. and R.M. v. International Protection Appeals Tribunal and Minister for Justice [2021] IEHC 817: Ferriter J. emphasised that s.31 requires effectiveness, not evidence that a state is merely trying; and criticised reliance on selectively extracted COI. The judgment also cited R (Atkinson) v. Secretary of State for the Home Department [2004] all England Reports (D) 14 for the idea that “incompetent best efforts” are not enough. O’Donnell J. used F.M. and R.M. by analogy: “improvement” or “effort” cannot be elided with the statutory threshold of “effective” protection.

v) Lawful preference between COI sources; judicial restraint

The Court contrasted the above with cases where the Tribunal’s COI methodology was upheld because it transparently selected the most specific, standards-compliant material and explained why:

  • X.S. and J.T. v. IPAT and Minister for Justice and Equality [2022] IEHC 100: Ferriter J. upheld a “cogently reasoned approach” where the Tribunal treated COI as mixed and rationally preferred a specific report (“CEDOCA report”) for blood-feud issues.
  • W.P.L. and BP v. The Minister for Justice and Equality, The International Protection Appeals Tribunal Ireland and The Attorney General [2024] IEHC 184: Phelan J. reiterated that courts should not interfere with conclusions open to a properly directed Tribunal; state protection must be effective, not perfect; and general crime levels do not necessarily negate protection where reasonable steps are being taken.

O’Donnell J. located the present case firmly within this framework: the Court’s role is not to re-decide protection, but to test whether the Tribunal’s conclusion was lawfully open on the COI.

vi) Broader international/EU context for gender-based persecution (contextual, not dispositive)

Although the applicant relied heavily on EU law and gender-based persecution jurisprudence, the Court noted that the Tribunal largely accepted the persecution narrative and Convention nexus, so the decisive defect lay elsewhere (state protection/COI). Still, the Court carefully set out:

  • WS v. Intervyuirasht organ na Darzhavna agentsia za bezhantsite pri Ministerskia savet C- 621/ 21, ECLI:EU:C:2024:47, 16 January 2024 (W.S.): the CJEU emphasised vigilant, individualised assessment, and the importance of gender-relevant COI (legal status of women, prevalence/incidence of violence, practical protection, penalties, risks on return). It also interpreted Directive 2011/95/EU consistently with CEDAW and the Istanbul Convention.
  • Staatssecretaris van Veiligheid en Justitie (Political opinions in the host Member State), C-151/22, EU:C:2023:688 (cited within W.S.).
  • AH and FN v. Bundesamt fur Fremdenwesen und Asyl, cases C-608/22 and S-609/22, ECLI:EU:C:2024:828 (reiteration of W.S. principles).
  • Gaidukevich v Georgia (Application No. 38650/18, 15 June 2023) was referenced by the applicant on the dynamics of domestic violence; the Court did not ultimately decide the case on that basis.

B) Legal Reasoning

i) What the Court accepted about the Tribunal’s approach

  • The Tribunal correctly identified s.31 as the governing test and referenced the four-part formulation from B.C. v. IPAT [2019] IEHC 763.
  • The Tribunal accepted credibility, the occurrence of persecution, a well-founded fear, and the Convention nexus (particular social group).
  • The Court rejected the “reasons” challenge: the decision was intelligible even if wrong in law on the COI/protection conclusion.
  • The Court was not persuaded there was the kind of selective COI use condemned in D.V.T.S. v. The Minister for Justice [2008] 3 IR 476.

ii) The decisive error: “improvements” treated as “effectiveness”

The Tribunal described the COI as presenting a “mixed picture” but then drew a firm conclusion of effective protection, emphasising developments (training, systems, advocates, electronic surveillance/communication) and a general “impression” of progress.

The High Court held that this reasoning failed at two levels:

  1. Rational connection: even allowing for the “not perfect but reasonable” standard, the Court found “very little” in the COI that plainly supported a finding that domestic violence state protection in Georgia is effective, with the general tenor indicating inadequacy. As in B.A. v. IPAT [2020] IEHC 589, the positive legal conclusion did not coherently follow from the evidential account.
  2. Misdirection in law: drawing on the logic of F.M. and R.M. v. International Protection Appeals Tribunal and Minister for Justice [2021] IEHC 817, the Court held that the Tribunal “erroneously elided improvements with effectiveness”. Section 31 is not satisfied by evidence of reform momentum or governmental endeavour; the question is whether protection has reached the statutory threshold of being effective and non-temporary, grounded in an effective system for detection, prosecution, and punishment.

iii) Judicial review boundary respected

A central theme is that quashing did not arise because the High Court preferred a different view of Georgian COI. Instead, the Court framed the defect as one of legality: the conclusion was not reasonably open on a fair reading of the COI and reflected the wrong legal yardstick (improvement ≠ effectiveness). This tracks the restraint emphasised in X.S. and J.T. v. IPAT and Minister for Justice and Equality [2022] IEHC 100 and W.P.L. and BP v. The Minister for Justice and Equality, The International Protection Appeals Tribunal Ireland and The Attorney General [2024] IEHC 184.

C) Impact

  • Sharper discipline in s.31 findings: Tribunals must demonstrate (not merely assert) how COI establishes “effective” protection. A narrative of reforms, training, or new laws will not suffice unless it supports the statutory conclusion in practice.
  • COI “mixed picture” cannot be used as a bridge to certainty without analysis: where COI contains both progress and persistent inadequacy, the Tribunal must show why the statutory threshold is met, not simply note improvement and proceed to effectiveness.
  • Domestic/gender-based violence claims: the decision underscores that where persecution and nexus are accepted, state protection becomes the crucial battleground; the COI evaluation must be especially careful because “improving frameworks” may coexist with ineffective frontline protection.
  • Future litigation focus: applicants will likely frame challenges around the “improvements vs effectiveness” distinction and the rational-connection test between the COI record and the s.31 conclusion, rather than reasons-form challenges.

4) Complex Concepts Simplified

Country-of-Origin Information (COI)
Objective material (e.g., NGO reports, government human-rights reports, international monitoring) used to assess conditions in the applicant’s home country—especially whether the state can protect against the feared harm.
State protection (s.31 of the 2015 Act)
Not a guarantee of safety. The question is whether the state provides effective, non-temporary protection through reasonable preventive steps and an effective legal system for detecting, prosecuting, and punishing persecutory/serious-harm acts—and whether the applicant can access it.
“Improvements” vs “effectiveness”
A state may pass new laws, train police, or create services (improvements), yet victims may still face inadequate enforcement, poor investigations, or institutional barriers. The legal test requires effectiveness, not progress.
Convention nexus / Particular social group
Refugee status requires persecution “for reasons of” a Convention ground. Here the Tribunal accepted that the applicant’s risk arose by reason of membership of a particular social group (women/victims of domestic violence in the relevant formulation).
Judicial review vs appeal
The High Court does not re-make the protection decision. It supervises legality: whether the Tribunal applied the correct legal test, considered relevant material fairly, and reached a conclusion that was reasonably open on the evidence.

5) Conclusion

M.G. (Georgia) v The International Protection Appeals Tribunal and Anor [2026] IEHC 151 reinforces a precise and practically important rule for international protection decision-making: evidence of legal and institutional reform in a country of origin cannot be treated as synonymous with “effective state protection” under s.31 of the 2015 Act.

The Tribunal may find COI “mixed” and may acknowledge improvements, but it must still demonstrate—by rational engagement with the COI as a whole—that the statutory threshold of effective, non-temporary protection is met. Where the COI’s tenor indicates persistent inadequacy and the decision leaps from “improvement” to “effectiveness,” the resulting finding is not reasonably open and will be quashed.