SGO in Care Proceedings: Proportionality Required and Threshold Typically Determined Before Permanently Placing a Child with a Foster Carer

1. Introduction

This appeal concerned a child, E (aged six), whose early life involved serious safeguarding concerns linked to the mother’s postnatal mental illness and subsequent diagnosis of bipolar disorder. Earlier care proceedings ended with an SGO to the maternal grandmother (the MGM) and an intention that the mother, MGM and E would live together.

In March 2024, following renewed concerns and a contested removal, E returned to public law proceedings and was placed in foster care under an interim care order. By the final hearing (January 2026), E had been with a second foster carer for nearly 18 months and had frequent supervised contact with the mother and MGM. The Local Authority’s final plan—supported by the children’s guardian—was for an SGO to the foster carer. The mother and MGM sought return to their (sole or joint) care, indicating willingness to accept a supervision order.

The key appellate issue was methodological: what legal approach must a judge take when making an SGO to a foster carer at the end of care proceedings—particularly where the alternative is reunification?

2. Summary of the judgment

The Court of Appeal (Moylan LJ giving the lead judgment; Singh LJ and Phillips LJ agreeing) allowed the appeals and remitted the matter for rehearing.

3. Analysis

3.1 Statutory context: why an SGO in care proceedings can be “intrusive”

The court mapped the Children Act 1989 structure: Part I (welfare), Part II (including SGOs), Part IV (care/supervision). Although an SGO is not made under s.31, its legal effect is substantial. Under s.14C(1), a special guardian may exercise parental responsibility to the exclusion of others with parental responsibility (save another special guardian). That feature—coupled with the “permanence” purpose of special guardianship—means that an SGO to a foster carer can amount to a major state-facilitated interference with family life.

The court also noted the procedural reality: a Local Authority cannot itself apply for an SGO (s.14A(5)), but can pursue it through its care plan, with the court empowered to make it even without an application (s.14A(6)).

3.2 The central correction: welfare is not enough—proportionality is required

The first-instance judge stated he needed only to decide “which course will best serve” welfare, and he did not address proportionality. The Court of Appeal held this was a legal misdirection in the context of care proceedings that may produce an effectively permanent separation from the birth family.

Moylan LJ anchored the requirement in Supreme Court authority, especially In re B (A Child) (Care Proceedings: Threshold Criteria) [2013] UKSC 33, [2013] 1 WLR 1911 (“Re B [2013]”) and In re H-W (Children) [2022] UKSC 17, [2022] 1 WLR 3243 (“Re H-W”). The key move was to treat an SGO-to-foster-carer outcome as belonging to the same “intrusive” family-separation category as long-term foster care under care orders—so that Re B/Re H-W proportionality principles apply mutatis mutandis.

The court’s approach “needs to reflect” the significant interference; the decision is “more than a simple welfare balancing exercise.”

3.3 Threshold: not always legally necessary, but typically necessary in this scenario

A distinctive aspect of this judgment is its treatment of s.31(2) threshold. Moylan LJ accepted the formal point: threshold is a statutory gateway for making care/supervision orders, and the court did not make such an order. However, he held that, typically, the court should still determine threshold when the case is the conclusion of care proceedings and the contested outcomes include a non-family SGO and possible reunification (potentially with supervision).

Three reasons were given:

• Option appraisal requires it: because a supervision order was part of the family’s proposals, the court could not responsibly evaluate that option without deciding whether the s.31(2) criteria were met; otherwise, the judge risks implicitly pre-determining the outcome.
• Threshold is a safeguard against unwarranted state interference: drawing on In re J (Children) (Care Proceedings: Threshold Criteria) [2013] UKSC 9, [2013] 1 AC 680 (“Re J”) and Baroness Hale’s explanation that the threshold “restrict[s] compulsory intervention to cases which genuinely warrant it”.
• Factual discipline and risk analysis: adopting a threshold-style framework promotes clear findings about what harm the child must be protected from—supporting robust welfare/risk reasoning.

3.4 The precedents cited: what they contributed

• In re J (Children) (Care Proceedings: Threshold Criteria) [2013] UKSC 9, [2013] 1 AC 680 (“Re J”): Used to explain the constitutional/human rights function of threshold as a protection for families against unjustified compulsory intervention, while permitting child protection where warranted.
• In re B (A Child) (Care Proceedings: Threshold Criteria) [2013] UKSC 33, [2013] 1 WLR 1911 (“Re B [2013]”): Cited for the necessity/proportionality framework and for the warning that the reason for crossing threshold matters—because “it is not sufficient that the child would be better off in another family”.
• In re H-W (Children) [2022] UKSC 17, [2022] 1 WLR 3243 (“Re H-W”): The pivotal authority extending (or reaffirming the extension of) Re B proportionality beyond adoption plans to other long-term separations (care orders with long-term foster care). Moylan LJ treated SGO-to-foster-carer as engaging the same logic.
• Re B (A Child) (Adequacy of Reasons) [2022] EWCA Civ 407; [2022] 4 WLR 42 (“Re B [2022]”): Relied on for the importance of clear threshold findings and warnings against prematurely ruling out options without proper analysis. Also emphasised that disputed matters significant to the decision must be decided.
• L-G (Children: Risk Assessment) [2025] EWCA Civ 60, [2025] 1 FCR 695: Cited to criticise “risk” reasoning that is not properly analysed (e.g., how likely is relapse; what are the consequences; how can it be managed).
• K-H (Children)(Care Orders: Proportionality) [2025] EWCA Civ 1368, [2025] 4 WLR 116: Used to link the assessment of likelihood of harm under s.31 threshold and under the welfare checklist, indicating methodological consistency.
• Re N (A Child: Placement Order: Proportionality) [2025] EWCA Civ 154 and F (Children: Placement Orders: Proportionality) [2026] EWCA Civ 990: Cited for structured questions that courts should ask when contemplating removal from family life, culminating in whether interference with Article 8 rights is necessary and proportionate.
• Re T (Children: Risk Assessment) [2025] EWCA Civ 93; [2025] 4 WLR 97; [2025] 2 F.L.R. 263: Reinforced a structured, question-led risk analysis (what harm, how likely, consequences, mitigation).
• Re A (Special Guardianship: Competing Applicants) [2018] EWCA Civ 2240, [2019] 1 FLR 687: A direct signpost that an SGO decision in care proceedings can require an explicit proportionality evaluation of state interference in Article 8 rights.
• Re S (Adoption Order or Special Guardianship Order) [2007] EWCA Civ 54, [2007] 1 FLR 819 and In re M [2021] EWCA Civ 442, [2022] Fam 53: Used to underline special guardianship’s “permanence” function in care planning.
• In re G (Children) (Residence: Same-Sex Partner) [2006] UKHL 43, [2006] 1 WLR: Referred to in argument to show that while there is no formal “natural parent presumption”, biological/familial relationships remain a significant factor in welfare evaluation; the appeal court did not decide this point as a standalone error, but it framed the critique of an over-simplified welfare balance.
• In re F and another (Children) [2021] EWCA Civ 622, [2022] Fam 78: Cited in submissions (especially for the proposition that an SGO can be highly significant and potentially “akin” to adoption in its practical permanence, albeit without extinguishing parental responsibility).

3.5 Legal reasoning: what the Court of Appeal required the judge to do

The Court of Appeal required a public-law removal decision—even if implemented via Part II (SGO)—to be justified through a structured evaluation, including:

• Findings and factual foundation: determine disputed matters underpinning the asserted harm/risk and set out how threshold (typically) is met (what harm, attributable to what).
• Risk analysis that is more than assertion: quantify/assess likelihood and consequences of future harm (e.g., relapse), and consider what measures could reduce/manage risk (support, monitoring, relapse plan, supervision).
• Comparative welfare evaluation: weigh realistic options (return to mother; return to MGM; joint care; foster carer under SGO; care order), with clear reasons for rejecting less interventionist options.
• Necessity and proportionality: explain why the chosen arrangement is the least interventionist measure consistent with the child’s welfare and protection from harm, in Article 8 terms.

The judgment also flagged, without needing to decide them, additional concerns that were capable of affecting outcome on rehearing: the staleness of key assessments of the mother (over 18 months old by final hearing), the Local Authority’s apparent early “ruling out” of the mother despite professional evidence that her difficulties might be addressable, and the need for clear findings on conflict between the mother and MGM and how that conflict interacted with risk.

3.6 Impact: likely effects on future practice

Judicial approach: The decision is a clear appellate steer that courts must treat an SGO-to-foster-carer outcome in care proceedings as engaging the same family-separation gravity as long-term foster care under a care order, requiring explicit proportionality reasoning.

Local Authority and guardian evidence: Care plans proposing SGOs to foster carers should anticipate structured proportionality scrutiny and be evidenced accordingly: up-to-date assessments, clear articulation of the harm feared, the probability and impact of that harm, and why support/monitoring would not sufficiently mitigate risk in a family placement.

Threshold discipline even when not making s.31 orders: While Moylan LJ preserved flexibility (“typically” rather than “always”), practitioners should expect that courts will often require a threshold determination (or threshold-like clarity) before endorsing a plan that effectively places the child permanently outside the birth family under an SGO.

Option of supervision order: Where a supervision order is advanced as part of a reunification plan, this case underscores that the court must squarely address it and cannot sidestep it by reasoning backwards from an intended non-s.31 outcome.

4. Complex concepts simplified

• “Threshold” (s.31(2) Children Act 1989): The statutory gateway for state-compelled child protection orders (care/supervision). It asks whether the child has suffered or is likely to suffer significant harm, and whether that is attributable to parental care (or lack of control). It protects families from the state intervening just because an alternative placement might look “better”.
• “Proportionality” / “necessity” (Article 8 ECHR): Even if risks exist, the court must choose the least interventionist order that will protect the child. Permanent or long-term separation from the family must be justified as necessary to meet a pressing social need, and proportionate to the harm found or feared.
• Why SGOs can be highly interventionist: An SGO gives the special guardian parental responsibility and practical control over decision-making to the exclusion of others with parental responsibility. When granted to a foster carer at the end of care proceedings, it can function as a long-term “settlement” outside the birth family.
• “Welfare checklist” (s.1(3) Children Act 1989): A list of factors (needs, effect of change, harm, capability, etc.) guiding what is best for the child. This case stresses that in public law removal contexts, the checklist must be integrated with proportionality reasoning, not used as a substitute for it.

5. Conclusion

Re E (A Child) (Special Guardianship Order: Public Law Proceedings) [2026] EWCA Civ 1167 establishes (and consolidates) that when care proceedings end with a proposed SGO to a foster carer—especially where reunification is the alternative—the court must: (i) treat the decision as a serious, intrusive interference with family life; (ii) conduct a structured necessity and proportionality analysis under Article 8; and (iii) typically determine s.31(2) threshold (and make clear findings on disputed facts) to ground the risk/harm evaluation and the comparative appraisal of options.

The consequence in this case was procedural but significant: the SGO was set aside and the matter remitted, underscoring that a legally incomplete methodology—however carefully expressed in welfare terms—cannot stand where the outcome is long-term separation from the child’s family.