Wells sharing is subject to no principle confining it to a last resort or to a minority element of an award. That is the holding in Culligan v Rosemin-Culligan [2026] EWCA Civ 948, handed down on 24 July 2026. The Court of Appeal allowed a husband’s appeal and increased the wife’s share of an illiquid private company holding from 30 per cent to 50 per cent.
What the judge below had ordered
The marriage lasted 40 years. MacDonald J made a final financial remedy order in September 2025 dividing the parties’ capital equally by value, giving each approximately £13.7m net.
The difficulty lay in the structure rather than the headline. Of the husband’s share, 70 per cent comprised illiquid assets; of the wife’s, 30 per cent. The illiquid element in both cases was shares in Colendi Holdings Limited, valued at £19m gross. Those shares came from the husband’s holding in SETL Limited, itself built from the proceeds of an early Bitcoin investment. Colendi acquired SETL in a share swap in January 2023.
MacDonald J found it fair for the husband to bear a larger share of the consequences of that transaction. The basis was that he had converted a substantial matrimonial asset into an illiquid minority interest without consulting the wife. He fixed the extent of Wells sharing at 30 per cent. Wells sharing, derived from Wells v Wells [2002] EWCA Civ 476, is the practice of dividing risk-laden assets in specie.
Where did the judge go wrong?
Moylan LJ gave the lead judgment, with which Coulson and Arnold LJJ agreed. The judge had been led into error by treating Versteegh v Versteegh [2018] EWCA Civ 1050 as establishing that Wells sharing should be a “last resort” comprising only a “minority element” of an award.
That formulation does not come from the ratio of the Court of Appeal in Versteegh. It comes from a passage of Mostyn J which Lewison LJ quoted without endorsing it. No such principle exists. Moylan LJ was equally clear that there is no principle pulling the other way either, and that Wells sharing is not to be regarded as the default.
What remains is the approach in Martin v Martin [2018] EWCA Civ 2866. It is a broad evaluative exercise directed at a fair balance of risk and illiquidity between the parties, with no percentage ceiling to work back from.
The substantive findings did not survive either
Two evidential conclusions fell away on appeal. The finding that the husband had failed to consult the wife before the SETL sale sat inconsistently with the judge’s own earlier rejection of that same conduct allegation. The evidence also indicated the sale would have gone ahead regardless, SETL’s position being precarious without the Colendi transaction.
Nor did the evidence support treating the Colendi shares as more illiquid or riskier than the SETL shareholding they replaced. If anything the swap had improved the prospect of realising value. The premise on which the unequal risk allocation rested therefore did not hold.
What the Court of Appeal substituted
The parties’ housing needs and earning capacities were broadly equivalent. Giving limited weight to the husband’s continued involvement with Colendi, a shareholding of only 3.6 per cent, the court concluded the shares should be shared equally.
The order was varied to increase the wife’s contingent entitlement from 30 to 50 per cent of the value received from the shares. The former matrimonial home is to be sold, with the husband taking 40.4 per cent of the net proceeds to balance the adjustment. That percentage applies irrespective of the eventual sale price. Market risk therefore falls on both parties in fixed proportions rather than being argued about later.
The practical effect is to remove a ceiling that practitioners had been arguing to. Where the pot is dominated by shares in a private company, a party seeking substantial in specie sharing no longer has to overcome a supposed rule that it should be exceptional and small. The schedule needs to show who carries the risk of realisation and why. A judge limiting the extent of Wells sharing now needs a reason grounded in the evidence rather than in a formulation lifted from Versteegh.
How should schedules be drafted now?
The judgment rewards attention to the comparative exercise. The court did not ask in the abstract whether the Colendi shares were illiquid. It asked whether they were more illiquid than what they replaced, and found they were not. Where an asset has changed form during the marriage, the risk analysis needs to run against the predecessor asset rather than against cash.
Two other points carry across. A conduct finding that has been rejected at one stage cannot be reintroduced as a reason for unequal risk allocation at another. And fixing a percentage of net proceeds that holds whatever the sale achieves is a device worth borrowing whenever a balancing payment depends on a future sale.
The family bench has been busy this summer. See also our reports on the new regulated expert requirement under rule 25.5A and on the Ministry of Justice consultation on weddings law reform.
The decision is reported at [2026] WLR(D) 419. The ICLR WLR Daily report carries the headnote, and the full judgment is available through Find Case Law.