Occupation orders and interim child arrangements after KM v LN [2026] EWHC 2123 (Fam)

by Edward Lamb KC

A marriage ends acrimoniously. The parties live together. The money to rent somewhere else is sitting in a bank account. Neither parent will leave the home. There are serious allegations yet to be determined. Into that stand-off the court inserts a “partial nesting” arrangement: the children stay put, both parents live there most of the week, and on alternate weekends one of them decamps to a hotel so the other can have the children alone. Nobody agreed to it. It is imposed, labelled temporary, and then renewed hearing after hearing. Ten months later it is still running.

That was the position in KM v LN when HHJ Middleton-Roy, sitting in the Central Family Court on 3 July 2026, decided that enough was enough. He made an occupation order under s33 of the Family Law Act 1996 requiring the father to leave within twelve days and excluding him for six months, and he revised the interim child arrangements so that the children would spend alternate weekends and a couple of midweek slots with him. The final financial remedy hearing was seventeen days away; the final children hearing was two months off. The father appealed, and on 7 August 2026 Poole J dismissed the appeal in a judgment that is now a leading modern authority on what happens when an occupation order and a child arrangements order are “inextricably linked”.

The wrong test, or the wrong order?

The father’s single appeal ground for which permission was given, was simple. The occupation order would inevitably end the nesting arrangement. Ending the nesting arrangement meant varying the child arrangements. Varying the child arrangements engages s1 of the Children Act 1989. So, the argument ran, the judge was wrong to decide the case “only through the prism of section 33” rather than also through the prism of the paramountcy principle and the welfare checklist, and in doing so he deprived himself of the “global, holistic and multi-faceted analysis” the children deserved [§9.1].

Poole J rejected the premise but accepted a refined version of the conclusion. Neither the paramountcy principle nor the welfare checklist applies to the decision whether to make an occupation order under s33(6) or s33(7). There was no error of law in declining to apply them [§30]. But that was not the end of it. Where occupation of the home and the child arrangements are closely linked, a judge “was bound to consider at least the shape of the reviewed child arrangements in the event of F being excluded” [§31]. The route into that consideration is the statute itself: s33(6)(c) requires the court to weigh the likely effect of any order, or of no order, on the health, safety or well-being of the parties and any relevant child. If the order will change who lives where and how often the children see each parent, the court cannot assess that effect without first forming a view of what the post-exclusion arrangements, made in the children’s best interests, would look like.

Poole J was alive to the obvious objection and met it head-on: “This is not a circular exercise” [§32]. The court works out what child arrangements would serve the children if an order were made, and then weighs those arrangements in the s33(6) balance (or, where significant harm is in play, under s33(7)). The welfare analysis informs the occupation decision; it does not replace it. The father’s suggestion that excluding one parent “did not mandate” the particular arrangements then ordered was accepted in principle, since a judge might in theory have ordered a 50/50 split, but it did not help him, because the judge had considered and rejected equal time on the evidence [§32, §45].

The sting for the appellant was in the sequencing. The notes of the ex tempore judgment showed that the judge turned to the revised child arrangements only after he had decided to exclude the father. That was enough to give ground 1 a real prospect of success [§33]. Reading the judgment as a whole, in the manner Lord Hoffmann demanded in Piglowska and Munby P reiterated in Re F, Poole J was satisfied that the judge knew exactly what would follow from exclusion, because he was the one about to decide it, and that he had the children’s reduced time with their father in contemplation when he made the order. “The order of reasoning within his judgment did not affect the outcome” [§34]. It would have been better done the other way round; but it was not wrong.

What else the case decides

Three further points are of use to the family lawyer. First, s33(6) can be exercised, and a parent excluded from a home s/he co-owns, with conduct expressly treated as neutral. There had been no fact-finding and no admissions. The judge did not rely on either parent’s conduct “to tip the scales” [§41], and Poole J thought that “perhaps wise”. The exclusion rested instead on the “virtual certainty” that continued cohabitation would harm the mother and the children, set against the risks of change. Building on Re L (Occupation Order) [2012] EWCA Civ 721, the court confirmed that harm may flow from the parents’ presence together under one roof rather than from anything either of them has done, and that an order may be made “even in the absence of a finding that he was guilty of any reprehensible conduct” [§44].

Second, imminent final hearings do not oblige the court to wait. The timing was a relevant circumstance under s33(6) and the judge was found to have weighed it, but it “cannot reasonably be argued” that he was bound to stay the application; an occupation order may be justified “even if it will have effect only for a few days” [§29]. Allowing a damaging situation to run for a further two months, with the children on school holiday for most of it, would itself cause harm [§37].

Third, the familiar submission that exclusion will “paint him as the culprit” in the children’s eyes was acknowledged as a genuine risk, but “one factor” that “can be managed” and which was outweighed by the children’s need for breathing space from the conflict [§48]. On the facts, which parent should go was answered by the disparity of resources, the fact that the father had not himself applied for an occupation order, and the wishes of the twelve-year-old [§39–40, §43].

Practical lessons

For applicants: if an occupation order will dismantle an existing child arrangements order, come to court with a worked proposal for the day after the order is made. Tell the judge what the children’s week will look like, why that serves their welfare, and invite the court to weigh it under s33(6)(c) before, not after, the court decides on exclusion. For respondents: the Children Act is a shield here, not a sword. Arguing that the welfare checklist governs s33 will fail; arguing that the judge never considered what the children’s lives would look like post-exclusion may succeed, but only if the judgment really is silent on it.

Finally, KM v LN is a quiet verdict on imposed nesting. Poole J observed that high mutual hostility “was not fertile ground” for an order requiring warring parents to share a home with their children for most of the week [§44]. When the evidence suggests that the cohabitation itself is the most harmful factor in the children’s lives, s33 gives the court the tools to remove a harmful nesting arrangement, and this judgment explains how to use them.