Executive Summary
New Zealand is this series' clearest existence proof: a Hague system that is both fast and fair. In the 2021 global study it recorded a 71% return rate (17 of 24 applications, 16 of them by judicial order) and resolved cases in about 135 days end-to-end — well below the 207-day global mean, with a full appellate structure intact. And its small Supreme Court produced Secretary for Justice v HJ (2006), one of the Commonwealth's leading statements on the Convention's hardest moment: how a court should exercise its discretion once a child is already settled. The lesson is that fast and fair are not rivals — speed in the ordinary case is what earns a system the legitimacy to exercise open discretion in the extraordinary one. This is educational, not legal advice.
Introduction
This series has dissected slow systems at length — Türkiye's 384 days [#21], Mexico's pending files [#11], Brazil's appeals [#1]. Fairness, and usefulness, demand equal time for the systems that work. Israel's 83-day full cycle was one proof [#10]; Germany's 97-day courts another [#9]. The third — smaller, quieter, and doctrinally among the most influential — is New Zealand.
The 2021 global study's numbers for New Zealand read like a design specification for the treaty itself. Of 24 incoming return applications, 17 ended in the child's return — a 71% return rate, among the highest recorded for any active system. The Central Authority moved applications to court in 26 days on average, and the courts decided in 129; end to end, the whole system averaged 135 days. (The 26- and 129-day figures are stage averages drawn from a slightly different sub-sample and are not simply additive to the 135-day overall figure — but each stage is genuinely fast.) That 135-day average sits well below the global mean of 207 days , and is nearly three times faster than the slowest corridors, such as Türkiye's 384 [#21] — all with a full appellate structure available. Sixteen of those seventeen returns were judicial — this is not a system avoiding hard cases; it is a system deciding them quickly.
Yet New Zealand's greatest contribution to this field is not a statistic. It is a judgment — one that shaped how the common-law world treats the Convention's most delicate moment: the child who is already settled.
Legal Background: return, settlement, and the discretion question
Two points frame the case that follows. First, as throughout this series: a Hague return decides only which country's courts resolve custody, not who wins it. Return ≠ custody. Second, the Convention is not absolute. Article 12 provides that where more than a year has passed and the child is now settled in the new environment, a court is no longer bound to order return — a defence New Zealand enacts as section 106(1)(a) of the Care of Children Act 2004. Once such a defence is established, a hard question opens: does the Convention's strong pro-return policy still dominate — so that refusing return requires something "exceptional" — or is the court's discretion genuinely open? That question is the subject of Secretary for Justice v HJ.
What happened
Secretary for Justice v HJ reached New Zealand's newly created Supreme Court in 2006. Its shape is familiar from this series: children removed from Australia to New Zealand in early 2002; return proceedings not commenced until late 2003, nearly two years later; and by the time the courts faced the question, the children's lives had rooted in New Zealand and the settlement defence under s 106(1)(a) was squarely engaged [the arithmetic of articles #1, #5, #15].
The legal question was the one that splits the treaty's soul: once a defence like settlement is established, how should a court exercise its discretion — does the Convention's pro-return policy still dominate, such that refusal requires something "exceptional"? Or is the discretion genuinely open?
The Supreme Court held the discretion to be broad and unfettered — at large, with no "exceptional circumstances" hurdle. A court weighs the Convention's purposes — prompt return as the general interest of children as a class, deterrence of abduction, comity between member states — against the actual circumstances that established the defence and the welfare of the actual child before the court. Where settlement is real, the treaty's central promise (swift return before a new life takes root) has already failed by definition, and its policy weight diminishes accordingly. The appeal was dismissed; the children stayed.
That framework did not stay at the bottom of the Pacific. The following year, the House of Lords in Re M [article #5] articulated the same architecture for the United Kingdom — Baroness Hale holding that "in cases where a discretion arises from the terms of the Convention itself... the discretion is at large," and rejecting any additional test of exceptionality. Whether or not one court directly influenced the other, the Commonwealth's settlement-discretion doctrine crystallised in these two judgments within a year — and New Zealand's small Supreme Court got there first. Small systems can lead.
Why the machine is fast
New Zealand's speed is not geography's gift alone. Its components are legible and, mostly, copyable:
- The Convention is domestic statute. The Care of Children Act 2004 (ss 94–124) writes the Hague machinery directly into New Zealand family law — grounds, defences, discretion, timelines — so no court ever wonders how treaty and local law interact. Contrast Mexico's implementing-legislation gaps [#11].
- An operational, voluntary-first Central Authority. The Central Authority within the Ministry of Justice actively pursues voluntary return before and alongside litigation, and facilitates the judicial proceedings for the petitioning parent — the applicant abroad is not left to navigate a foreign system alone. The 26-day desk-to-court figure is what an intake office looks like when it sees itself as a pipeline, not a gate [contrast Spain's Article 27 outlier, #22].
- A unified, specialised family jurisdiction in a small country: the expertise-concentration effect Germany legislated deliberately [#9], New Zealand gets structurally.
- Island prevention synergy. As in the UK [#18], geography helps — every exit is a port, so measures against removal are more readily enforceable, which suppresses some of the caseload before it ever becomes litigation.
Honest caveats belong in the same paragraph. Twenty-four cases is a small denominator — single years swing. The Central Authority rejected 3 of 24 applications at intake — a 12.5% gatekeeping rate that deserves the same reasons-and-review transparency this series demanded of Spain [#22]. And two advantages — scale and isolation — do not travel: Germany proves the model scales up, but no policy transplant gives a landlocked state a coastline. What travels is the design: statute, active CA, concentrated courts, voluntary-first culture.
What This Shows About the Limits of the Hague Convention Alone
New Zealand shows that the Convention's text is not the variable that decides whether it works — the machinery around it is. The same treaty that runs 384 days in one country runs 135 in another; the difference is domestic statute, an active Central Authority, concentrated courts and a voluntary-first culture, none of which the Convention itself supplies. And the doctrinal point cuts the same way: New Zealand could afford an open settlement discretion precisely because its speed makes settlement rare. A fast system meets few settled children and can treat each as the exception it is; a slow system manufactures settled children by the calendar, then must decide whether the treaty means anything at all. The right order of reform is therefore always the same — fix the clock first, and the doctrine largely takes care of itself.
What Parents and Professionals Should Understand
For parents, the practical read — context, not legal advice — is that speed is your ally: in a fast system the settlement defence rarely gets off the ground, so the single most valuable thing anyone in a cross-border dispute can do is act early, before a new life takes root and HJ-style questions arise. For policymakers, New Zealand is the copyable existence proof: the design (Convention as domestic statute, an active voluntary-first Central Authority, a concentrated specialist jurisdiction) travels even where the geography does not, and the payoff is a system that is fast and keeps its full appellate safeguards. And the deepest professional lesson is that the small exemplars — Israel, Germany, New Zealand — are load-bearing for the whole field: they are what make the criticism of everyone else fair, because "it cannot be done" is refuted on three continents at once.
Limitations
New Zealand's 2021 figures rest on a small caseload (24 return applications), so single years can swing and the numbers indicate a pattern rather than a precise ranking; the 2021 data was also affected by the pandemic. The HJ and Re M holdings are summarised from the judgments and scholarly analysis. This is educational and not a substitute for advice from a qualified lawyer in the relevant jurisdiction.
Conclusion
New Zealand is the answer to the objection that runs beneath every slow-system chapter in this series — that fairness and speed pull against each other. Here is a jurisdiction that returns children in 135 days and wrote one of the Commonwealth's most careful statements of when not to return them at all. The two achievements are not in tension; they are the same achievement. A system fast enough to meet settlement rarely can afford to treat it seriously when it comes — and a system that treats every stage seriously earns the trust that makes its discretion legitimate. Fast, fair, and small: the existence proof the rest of the field is measured against.
Frequently Asked Questions
What makes New Zealand's Hague system fast? Four things: the Convention is written directly into domestic law (Care of Children Act 2004), the Central Authority actively pursues voluntary return and helps the applicant, family jurisdiction is concentrated and specialised, and island geography aids prevention. In 2021 it averaged about 135 days end-to-end, well below the 207-day global mean.
**What did Secretary for Justice v HJ decide?** That once a Hague defence such as settlement is established, a court's discretion whether to order return is "at large" — open, with no requirement of "exceptional circumstances." The court weighs the Convention's purposes against the child's actual circumstances and welfare. In HJ the children stayed in New Zealand.
Is a high return rate a good thing? It depends. A return decides forum, not custody, and the Convention has defences precisely because return is not always right. New Zealand's 71% is notable because it is achieved fast and mostly by judicial order — a sign of a system deciding hard cases promptly, not avoiding them.
Does the "settled child" defence mean a parent can win by delay? Not straightforwardly. It applies only after more than a year and requires genuine settlement, and even then the court has discretion. But it is why speed matters so much: the faster a case is brought and heard, the less likely settlement becomes a live issue.
References & sources
- Secretary for Justice (as NZ Central Authority) v HJ [2006] NZSC 97; [2007] 2 NZLR 289 — discretion framework; Courts of New Zealand case page: https://www.courtsofnz.govt.nz/cases/the-secretary-for-justice-as-the-new-zealand-central-authority-v-hj ; scholarly analysis: https://ojs.victoria.ac.nz/vuwlr/article/download/4768/4236
- Re M (Children) (Abduction: Rights of Custody) [2007] UKHL 55; [2008] 1 AC 1288 — "discretion at large" / no exceptionality [article #5 in this series]: https://www.incadat.com/en/case/937
- Care of Children Act 2004 (NZ), ss 94–124 (Hague implementation; s 106 grounds for refusal): https://www.legislation.govt.nz/act/public/2004/0090/latest/whole.html
- ICMEC, New Zealand — country profile (Central Authority practice): https://www.icmec.org/wp-content/uploads/2015/10/New-Zealand.pdf
- N. Lowe & V. Stephens, HCCH Prel. Doc. 19A (Fifth Statistical Study, 2021 data) — New Zealand data (¶69; ¶112; Annexes 4, 7–8): https://assets.hcch.net/docs/a75d7234-deb9-4764-be72-a4a9d87c8af7.pdf