Sum-of-the-parts value and the case for structural change

When the wrapper stops working. Why South African investment holding companies persistently trade below intrinsic value, and what portfolio concentration, conglomerate dynamics and structural change reveal.

A valuation puzzle commonly discussed in the South African corporate finance context – the discount to net asset value – is real, but not uniform. Apply it broadly across all listed companies and the argument becomes imprecise. The discount that is most structurally persistent sits specifically with listed investment holding companies: the gap between their market capitalization and the sum-of-the-parts (SOTP) value of their underlying portfolios. Unlike a straightforward undervaluation, this discount is not a market error; it reflects investor scepticism about the cost and efficiency of the holding company wrapper itself, although there have been exceptions (notably the PSG Group, approximately 10 years ago). That gap, and what drives it, is the subject of this article.

Why investment holding companies trade at SOTP discounts

So obvious to retail investors?
An investment holding company offers investors exposure to a portfolio of assets through a single listed security. In theory, the wrapper should trade at or near the aggregate value of its constituent parts. In practice, it rarely does, and the reasons are not always clear. Generally, investors price the holding company at a discount to compensate for capital gains tax (CGT) leakage on asset disposals executed inside the vehicle, the incremental income tax drag on investment returns flowing through the structure, and corporate overhead costs at the centre – which the market views sceptically unless the holding company can clearly demonstrate that the value of its strategic oversight and capital allocation capability outweighs those costs. The risk that management’s decisions at the group level may not align with what shareholders would choose to do themselves adds a further discount. Where the portfolio consists largely of other listed entities, this cost is particularly transparent: investors can see exactly what each position is worth at any moment, making the drag of the wrapper immediately quantifiable. Listed South African investment holding companies have historically traded at SOTP discounts that have proven persistent across market cycles, as demonstrated by Remgro, Sabvest, Zeder and, before its restructuring, PSG Group. The last decade has seen this scenario play out with most investment holding companies.

The concentration effect: When one investment dominates

The holding company discount is not static; it widens materially as portfolio concentration increases. When a holding company is genuinely diversified, investors are paying for access to a portfolio across multiple sectors and risk profiles that would be difficult to replicate individually. That diversification premium partially offsets the structural drag. But when one investment grows to represent 50%, 60% or more of total SOTP, that argument collapses. Investors are no longer paying for a portfolio – they are paying a fee to own a leveraged, less-liquid proxy for a single underlying asset. The wrapper has ceased to be a benefit and has become a cost, and the market responds by widening the discount accordingly. The holding company’s traded value drifts further from intrinsic worth, not because the underlying asset has declined, but because the structural justification for the vehicle has eroded.

PSG Group and Capitec: The definitive South African case study

PSG Group’s restructuring illustrates this dynamic clearly. PSG Group was, for most of its listed life, a genuinely diversified vehicle – comprising financial services, education, food and agri, and private equity – and that diversification justified the structure. It was also the incubator of businesses like Capitec, and a listed entity with reasonable liquidity.

Over time, Capitec grew exponentially until it represented approximately 60% of PSG Group’s SOTP. The diversification rationale had dissolved: for financial services-focused investors, holding PSG Group had become functionally equivalent to an indirect, cost-encumbered stake in Capitec, trading at a persistent 30% to 40% discount despite the excellent capital allocation shown by PSG Group over the years, in terms of diversification. Other smaller investments complicated matters, and brought uncertainty regarding visibility of factors not aligned to financial services. This discount persisted, despite Jannie Mouton – the founder and Chairman who had identified and backed Capitec, PSG Financial Services (formerly PSG Konsult) and Curro long before the market appreciated their value – remaining at the helm of PSG Group. The structural drag of the wrapper proved even stronger than the credibility of one of the country’s most respected capital allocators.

The PSG Group board drew the correct conclusion and, in September 2021, PSG Group unbundled its Capitec stake directly to shareholders as a first step. A broader restructuring followed in 2022, with the remaining investments – PSG Financial Services, Curro, Zeder, KAL Group, CA&S and a portion of its interests in Stadio, among others – similarly distributed to shareholders, and the holding company delisting thereafter. The market’s response to the announcement was immediate: PSG Group’s share price closed a significant portion of its longstanding SOTP discount, confirming that the wrapper had become the obstacle rather than the vehicle. Of course, there may also have been other considerations for the unbundling.

The same dynamic played out, on a larger canvas, at Naspers and its Amsterdam-listed subsidiary, Prosus. Tencent, acquired in 2001 for $34 million, had grown to represent more than 80% of Naspers’ intrinsic value. In 2019, Naspers unbundled its international internet assets into Prosus and listed it on Euronext Amsterdam – a move intended to give the portfolio greater visibility and a broader institutional investor base. The structure, however, introduced new complexity: Prosus traded at a discount to its own SOTP, and Naspers in turn traded at a discount to its c.57% stake in Prosus. Attempts to address the situation through a cross-holding buyback mechanism – under which Prosus used Tencent sale proceeds to repurchase Naspers shares, and Naspers repurchased Prosus shares – provided partial relief, but did not resolve the structural problem.

The lesson has been clear: adding layers does not cure concentration. Genuine value realisation requires either distributing the dominant asset directly to shareholders or redeploying capital at sufficient scale to rebalance the portfolio.

For M&A advisers and capital allocators, the structural SOTP discount creates a persistent opportunity. Two observations are particularly instructive:

• The signal is not the size of the discount, but its trajectory. A holding company at a stable 25% discount may be in structural equilibrium. One whose discount has expanded from 20% to 40% over 24 months, driven by the concentration of a single subsidiary, is signalling that the wrapper’s rationale has eroded – and that the market has taken notice.
• Unbundling is frequently more value-accretive than a take-private. Where underlying assets arelisted and liquid, distributing them directly to shareholders avoids the need to find a buyer willingto underwrite full SOTP at a single point in time (applicable where the underlying entity meets therequirements as a standalone business). PSG Group demonstrated that the market would closethe discount organically once the wrapper is removed – no third-party capital was required.

Conclusion

The SOTP discount in South African investment holding companies is a structural feature, not a temporary inefficiency. Its drivers – such as CGT leakage, overhead cost, conglomerate discount and, above all, portfolio concentration – are well understood and consistently priced by the market. The PSG Group/Capitec case was not exceptional; it is instructive. For the right holding company – one with genuine diversification, active capital allocation and a clear shareholder value proposition – the listed structure remains a compelling platform. However, the holding company wrapper has a natural lifespan, and when the conditions that justified its creation no longer hold, structural change is the rational response. Identifying the inflection point ahead of the market remains one of the more consequential judgements in South African corporate finance.

Bhargav Desai and Sibongakonke Kheswa are Corporate Financiers | PSG Capital

Sources:
Anchor Capital (2021). Update on the Naspers/Prosus Discount to NAV. anchorcapital.co.za.
Moneyweb (Mar 2022). PSG to Unbundle its Investments and Delist.
Moneyweb (Apr 2022). PSG Group Bids Farewell with Good Results.
Perpetua Investment Managers (Aug 2025). The Investment Case for Naspers and Prosus. perpetua.co.za.
PSG Group (Mar 2022). SENS Announcement: Proposed Restructuring and Delisting. JSE SENS.

This article first appeared in DealMakers, SA’s quarterly M&A publication.

DealMakers is SA’s M&A publication.
www.dealmakerssouthafrica.com