087 822 1639
087 822 1639
Trusts and divorce in South Africa can create difficult questions about whether trust assets are relevant to a matrimonial-property claim — a trust is not automatically protected from a divorce claim, and it is not automatically included in one either, since whether trust assets become relevant depends on the matrimonial property regime, the type of claim being made, and specific evidence about how the trust was actually controlled and used, not simply on the fact that a trust exists.
South African courts have repeatedly warned against both extremes: assuming a trust offers complete protection, and assuming any trust connected to a spouse can simply be treated as that spouse’s personal property.
This is a genuinely unsettled and actively developing area of law within high-net-worth divorce in South Africa.
A substantial line of High Court and Supreme Court of Appeal authority has developed over the past two decades, and two of the most recent Supreme Court of Appeal judgments — delivered five weeks apart in 2022 — have reached conclusions that legal scholars regard as being in direct tension with one another.
This article works through what is actually settled, what remains genuinely uncertain, and why the answer depends heavily on which matrimonial regime applies.
| Question | General position |
Are trust assets automatically excluded from divorce? | No. A trust is legally distinct from its founder, trustees and beneficiaries, but courts can look behind that separation in defined circumstances. |
Are trust assets automatically included if a spouse has some connection to the trust? | No. Being a trustee, founder or beneficiary is not, by itself, enough. |
Does the same test apply regardless of matrimonial regime? | No. The test differs significantly between marriages in community of property, redistribution claims, and accrual claims. |
Is control over the trust always required? | Not necessarily. Recent Supreme Court of Appeal authority suggests a trust can be pierced even without proof of the traditional “control” test, in defined circumstances. |
Is this settled law? | No. Two Supreme Court of Appeal judgments delivered five weeks apart in 2022 on the accrual context are in acknowledged tension, and academic commentary treats the position as genuinely unresolved. |
Can a spouse who is not a trust beneficiary challenge how it was run? | It depends on the matrimonial regime and the nature of the claim — this has been a live and contested issue across several of the leading cases. |
This is the single most important organising question, and it is the one that existing commentary on this topic most often skips.
The courts have been explicit that the test for whether trust assets can be reached in a divorce is not the same depending on whether the marriage is:
Treating these as interchangeable is one of the most common errors in this area — including, at times, by the courts themselves, as the case law below shows.
Where spouses are married in community of property, there is generally one joint estate. The question is whether trust assets should be treated as forming part of that joint estate.
WT and Others v KT (933/2013) [2015] ZASCA 9; 2015 (3) SA 574 (SCA) dealt with exactly this situation.
A wife counterclaimed that a trust holding the couple’s former home should be included in the joint estate, alleging both that her husband had deceived her and that the trust was his alter ego.
The Supreme Court of Appeal rejected both arguments. On the deceit claim, the court found no evidence whatsoever of any misrepresentation — the wife had, by her own admission in testimony, known about the trust throughout.
On the alter-ego claim, the court held that, on the facts of this particular matter, she had no standing to challenge how the trust was managed, because she was neither a beneficiary of the trust nor a third party who had transacted with it.
This standing limitation did not remain the last word — see REM v VM below, which addressed the same question specifically in the accrual context.
The court also drew an important distinction that has shaped the law since: it held that the discretion exercised in Badenhorst v Badenhorst (discussed below) arose specifically from section 7(3) of the Divorce Act, which applies only to redistribution claims for marriages out of community of property.
A court dealing with a marriage in community of property has no comparable discretion — its role is generally confined to dividing the joint estate equally, not to deciding whether outside assets should be added to it.
Where a marriage was concluded out of community of property before 1 November 1984, or otherwise qualifies for redistribution relief under section 7(3) of the Divorce Act, a different and older line of authority applies.
Badenhorst v Badenhorst [2005] ZASCA 116; 2006 (2) SA 255 (SCA) is the foundational case. A husband’s parents had transferred a farm into a trust rather than to him directly; he later formed a second trust, informing his wife it was to protect assets from creditors.
On divorce, she sought to have the trust assets included when the court exercised its wide redistribution discretion.
The Supreme Court of Appeal held that trust assets can be considered in a redistribution order where two things are proved: that the spouse in question controlled the trust — as a matter of fact, not merely as a matter of the trust deed’s formal terms — and that, but for the trust’s existence, that spouse would have acquired and owned the assets personally.
This became known as the control test, and it remains the settled starting point for redistribution claims specifically. Section 7(3)’s wide discretion is what gives courts the latitude to reach this result; that discretion does not exist in the same form outside the redistribution context, which is precisely the distinction WT v KT later drew on to reach the opposite result for a marriage in community of property.
This is the most complex area, and the one where existing commentary on this topic has most often been incomplete.
The early accrual cases. RP v DP and Others 2014 (6) SA 243 (ECP) was among the first cases to apply alter-ego reasoning specifically to an accrual claim, rather than a redistribution claim. The court found that a husband had used a trust as his alter ego and included its assets in the accrual calculation, applying reasoning drawn from Badenhorst despite the different statutory context.
REM v VM [2016] ZASCA 5; 2017 (3) SA 371 (SCA) then addressed, at Supreme Court of Appeal level, the standing question that WT v KT had raised two years earlier.
The court held that an aggrieved spouse’s ability to challenge a trust’s management should not be as narrowly confined as WT v KT suggested, at least in the accrual context.
The court set out what commentators describe as a two-stage test: first, that the trust is genuinely being used as the spouse’s alter ego (the control test); and second, separately, that this alter-ego trust was used fraudulently or dishonestly specifically to evade a proper accounting of that spouse’s accrual.
The 2022 conflict. PAF v SCF [2022] ZASCA 101; 2022 (6) SA 162 (SCA) then went further, and is widely regarded by academic commentary as a genuinely significant development.
The facts were striking: roughly three weeks before the divorce trial, a husband created a trust in the British Virgin Islands, and the very next day donated R2.2 million to it. His brother was the sole trustee; his minor daughter the sole beneficiary.
It was common cause that the trust was legitimately established and that the husband could never personally recover the donation — meaning the traditional “control” element of the Badenhorst test was not straightforwardly present.
The Supreme Court of Appeal nonetheless held, unanimously, that the trust’s veneer could be pierced. The court confirmed that the power to pierce a trust’s form derives from the common law generally, is not confined to the redistribution context, and can apply to accrual claims as well.
Critically, the court held that the Badenhorst control test is not the only route by which abuse of the trust form may be established in an accrual matter — the timing, purpose and circumstances of a trust’s creation and funding may themselves be highly material, even without proof of the husband personally controlling or benefiting from the trust going forward.
Five weeks later, a different panel of the same court decided MJK and Others v IIK [2022] ZASCA 116; 2023 (2) SA 158 (SCA) — also an accrual case, involving three trusts and a close corporation.
That court held that the Badenhorst control test applies only to redistribution claims, not to accrual claims, and declined to pierce the trust and close-corporation structures at issue, finding no proof of fraudulent intent.
Academic commentary — including a detailed case note in the South African Law Journal and analysis published by De Rebus — has specifically flagged that the MJK court’s engagement with PAF v SCF was superficial, and that the two judgments, decided weeks apart by differently constituted panels of the same court, are difficult to reconcile.
As at the date of this article, no further Supreme Court of Appeal authority has resolved the tension, and commentators describe the correct test for accrual claims as genuinely uncertain.
What this means practically: a spouse relying on trust assets to support or resist an accrual claim should expect that the outcome may depend significantly on the specific facts, the timing of the trust’s establishment relative to the divorce, and how the evidence is framed — not on a single, settled legal test that can be applied mechanically.
Wealth structures sometimes combine a trust with a close corporation or company, adding a further layer of separation between a spouse and the underlying assets.
MJK and Others v IIK, discussed above, also raised a jurisdictional question specific to structures spanning multiple entities and, in that case, a foreign trust — namely, whether a South African court has jurisdiction to make findings about a trust established under foreign law.
The court’s approach confirms that combining a trust with a corporate entity does not simplify the analysis; if anything, each layer requires its own separate consideration of ownership, control and the applicable legal framework.
See Divorce and Business Ownership in South Africa for a fuller treatment of company and corporate structures in divorce.
Separately from the alter-ego and control-test analysis, a trust can also be attacked on the basis that it never validly came into existence at all — a sham trust.
This is a different and more fundamental challenge than “piercing” a validly established trust. A sham trust is one where the founder and trustees never genuinely intended to create a real trust relationship in the first place.
Courts assessing this question have looked at factors including whether trust assets are kept genuinely separate from the founder’s or trustee’s personal affairs, whether the trust’s stated purpose is authentic, whether the trust is managed independently of the founder’s personal wishes, and whether the founder retains unilateral power over trustees or beneficiaries inconsistent with genuine trust administration.
A sham-trust challenge is fundamentally different from an alter-ego or abuse-of-trust-form argument. As Van Zyl NNO & another v Kaye NO 2014 (4) SA 452 (WCC) explains, proving a trust is a sham concerns whether a genuine trust relationship was ever created at all — whether its apparent creation was merely simulated — which is a different question from “going behind” the form of a trust that was validly established but has since been misused.
Maladministration of a genuine trust does not, on its own, make it a sham. If the requirements for a valid trust were never genuinely satisfied, the supposed trust arrangement may be disregarded, and the proprietary consequences then depend on the actual underlying ownership and transactions proved on the evidence.
Depending on the pleadings and relief sought, the status and legal effect of the trust may need to be determined as part of, or alongside, the matrimonial-property dispute itself.
See Financial Disclosure in High-Net-Worth Divorce for the disclosure obligations that typically arise where a trust forms part of the estate.
Several of the cases above draw on foundational principles about how trusts must be governed, established well before any of these divorce-specific developments.
Land and Agricultural Bank of South Africa v Parker and Others 2005 (2) SA 77 (SCA) — a commercial case, not a divorce case — remains the leading authority on the basic separation the law expects between control of a trust (held by trustees) and enjoyment of its benefits (held by beneficiaries).
The court held that trustees must generally act jointly, that outsiders dealing with a trust cannot rely on the apparent authority of a single trustee the way they might with a company, and that informal, one-person-dominated administration of a family trust undermines its legal legitimacy.
Several of the divorce-specific judgments discussed above draw directly on this reasoning when assessing whether a trust has been properly, or improperly, administered.
Drawing the case law together, several recurring factors appear across the different tests, even though the precise legal threshold differs by context:
No single factor is decisive on its own. Courts have repeatedly emphasised that these questions are fact-intensive and cannot be resolved by assuming that any trust connected to a spouse is fair game, or that any properly documented trust is automatically beyond scrutiny.
When the property is put into a trust, the cause for the transfer is usually either a donation or a sale.
If the transfer of assets into a trust is a donation, donation taxes must be paid accordingly.
In the event of a sale, the purchase is frequently reported as a loan account in favour of the seller.
In actuality, assets are frequently transferred into trusts with little explanation.
Without proof of a donation or purchase consideration, such a transfer would constitute a loan account, which is an asset in the transferor’s estate.
If you are trying to include trust assets in a divorce claim:
If you are seeking to protect a genuinely independent trust:
Where a trust forms part of a wider coordinated wealth structure, see Family Offices and Divorce in South Africa for a related treatment.
Potentially, but the answer depends on your matrimonial property regime and the type of claim being made. The tests differ significantly between community of property, redistribution and accrual claims, and the accrual-context test is currently unsettled at Supreme Court of Appeal level.
Usually, yes, under the traditional control test established in Badenhorst. However, PAF v SCF suggests that in some circumstances a trust’s veneer can be pierced even without proof of ongoing control, where the trust’s creation and use amount to an abuse of the trust form.
It depends on the matrimonial regime and claim involved. WT v KT held that a spouse in a community-of-property marriage generally cannot, absent standing as a beneficiary or third party. REM v VM took a different approach in the accrual context.
Not automatically, but timing is a significant factor courts consider. PAF v SCF involved a trust created roughly three weeks before trial, and the timing was central to the court’s reasoning.
A sham trust is treated as never having validly existed at all. An alter-ego trust is treated as validly established, but its assets may still be considered relevant to a matrimonial claim because of how the trust has actually been controlled and used. The legal consequences and tests differ between the two.
No, particularly regarding accrual claims. Two Supreme Court of Appeal judgments delivered five weeks apart in 2022 — PAF v SCF and MJK v IIK — have reached positions that legal academics regard as being in genuine, unresolved tension. MJK v IIK is often cited by its 2023 South African Law Reports reference, but both judgments were handed down in 2022.
Trusts occupy a genuinely complex position in South African divorce law, and that complexity has only deepened as courts have grappled with an increasing variety of fact patterns. A trust is neither an automatic shield against a spouse’s claims nor an asset that can simply be assumed to belong to whoever is connected to it. The correct starting point is always the matrimonial property regime, followed by a careful, evidence-based assessment of control, timing, and how the trust has genuinely been administered — recognising that, particularly for accrual claims, the applicable legal test is currently the subject of real and unresolved disagreement at the highest appellate level.
Author: Martin Vermaak, B.Proc, LLB
Attorney of the High Court of South Africa | Director, Martin Vermaak Attorneys | Over 20 years’ experience in divorce and family law
Disclaimer
This information has been prepared for general educational purposes only and is not intended to constitute legal, financial, tax or other professional advice. South African family law is fact-specific, and the applicable legal position may depend on the circumstances of the individual matter. Readers should obtain independent professional advice appropriate to their circumstances before making legal, financial or other decisions.
– Executor And Estate Administration Services In South Africa
– Why You Must Have A Will (Pro’s And Con’s)
– How To Change Your Marital Regime
– The Myth Of Common Law Marriage In South Africa (Cohabitation)
– Antenuptial Contracts South Africa
– Executors Of A Deceased Estate
– EB (Born S) V ER (Born B) N.O. V And Others V KG V Minister Of Home Affairs And Others
– Offshore Assets And Your Will And Testament
– Do I Need An International Will
Copyright © 2026 Martin Vermaak Attorneys. All rights reserved.
Terms of Use Privacy Policy