JSE Limited: Can Africa's Oldest Exchange Survive the Companies It Made Rich?

I. Cold Open & Roadmap

On 2 March 2026, JSE Limited published a number it had never published before. Net profit after tax for the year ended 31 December 2025 came in at R1,071 million, up 16.7% from R918 million the year before — the first time in the exchange's 139-year existence that annual profit had crossed a billion rand.¹ The board raised the ordinary dividend 16.0% to 961 cents and layered a special dividend of 100 cents on top of it.¹ Return on equity climbed to 22.0% from 20.2%.¹

Two weeks later, the chief executive who had delivered that number handed over the keys and left. Leila Fourie, CEO since 2019, retired on 31 March 2026 — not pushed, not scandal-adjacent, not headhunted to a bigger job. She announced she intended to go sailing.² It is a rare thing in public markets: a CEO exiting on a record, at a time of her own choosing, with the balance sheet in better shape than she found it.

And yet the institution she left behind is presiding over something that looks a lot like a slow-motion evacuation. In the mid-1990s, the Johannesburg Stock Exchange hosted somewhere in the region of 850 listed companies. Today it hosts roughly 280.³ More than 500 companies have delisted over the past twenty-five years.³ The exchange has never been more profitable. The market it operates has never had fewer companies to trade.

That is the paradox this story has to resolve, and it is not a rhetorical one. It has a precise financial shape. An exchange makes money in two broadly different ways. It charges companies to be listed — a small, sticky, annuity-like revenue line. And it charges everyone else to trade, clear, settle, and get data on those companies — a much larger, much more cyclical revenue line that scales with activity, not with the number of names on the board. Those two things can move in opposite directions for a very long time. A shrinking universe of larger, more liquid companies can generate more trading revenue than a sprawling universe of illiquid small caps ever did. Until, at some point, it can't.

So the central question for anyone looking at JSE Limited as an investment is this: is this a royalty-like toll collector on South African and increasingly pan-African capital markets, mispriced at a fraction of what global exchange operators fetch? Or is it a franchise whose core product — public companies — is quietly disappearing out from under it, with the disappearance masked for now by a trading boom that management itself has told investors will not last?

The evidence for both readings is unusually good, which is what makes the case interesting rather than merely cheap.

Here is the route. First, how a noticeboard nailed up beside a gold rush became a for-profit company that lists on its own exchange and regulates itself as an issuer — a genuinely odd structural fact with live consequences. Then how JSE actually earns its money today, segment by segment, because the revenue engine is the precondition for every argument that follows. Then the fight that matters most: an active Competition Tribunal case that alleges JSE's dominance is sustained not by liquidity network effects but by rules — the single cleanest available test of whether this moat is earned or enforced. Then the listings exodus, its serious counter-narrative, and what the flow data actually shows. Then the leadership handover and FORGE 2031, the five-year plan a brand-new CEO has staked her tenure on. Then the valuation gap, the bull and bear cases stress-tested against the company's own history, and the small number of things worth watching from here.

Start with the noticeboard.

↑ Back to Top

II. From Gold Rush to Public Company: A Compressed History (1887–2006)

In 1886, a wandering prospector cracked open the Witwatersrand and found the largest gold deposit on earth. Within a year, the empty highveld had a town on it, the town had a hundred thousand speculators in it, and the speculators had a problem: they were trading mining claims and shares in mining syndicates out of tents, saloons and the open street, with no clearing, no settlement, and no reliable way to know what anything had last traded at.

On 8 November 1887, a London-born businessman named Benjamin Wollan opened the Johannesburg Exchange & Chambers Company — in practice a building with a noticeboard, where brokers could post bids and offers instead of shouting them. That is the founding myth, and it deserves exactly the weight a founding myth deserves: it explains the exchange's DNA and nothing about its economics. What matters is the pattern it established. From day one the JSE existed because a chaotic, capital-hungry extractive boom needed somewhere to price risk. Africa's oldest exchange was built as infrastructure for commodities. It is still, structurally, an exchange whose fortunes correlate heavily with the resources cycle and with foreign appetite for South African risk.

The century that followed was a slow build from noticeboard to real market. The first industrial listing arrived in 1897, when South African Breweries came to the board — the first sign the exchange could host something other than mines. The Stock Exchange Control Act of 1947 imposed the first proper statutory framework on what had been a self-governing club. The JSE joined the World Federation of Exchanges in 1963, formally taking its place among global peers. Electronic trading — the thing most Western readers assume was settled decades earlier — arrived only in the early 1990s, when the open-outcry floor closed and screens replaced it. That is a useful calibration: the market South Africans think of as ancient has only been electronic for about a generation.

But the hinge of the modern story is not any of those. It is a governance event, and it happened almost within living memory of current management.

The exchange that became a stock

For 118 years the JSE was a mutual — owned by its stockbroker members, run for their benefit, with a seat on the exchange functioning as both a licence to trade and an ownership claim. That structure has a well-understood pathology: the owners are also the customers, which makes it very hard to price aggressively, invest in technology that disintermediates members, or let outsiders in.

On 1 July 2005, the member-owned exchange demutualised and became JSE Limited — a company with shareholders rather than a club with members. In June 2006 it took the final step and listed on its own Main Board.⁴

Pause on how strange this is. JSE Limited is a company whose product is the regulation and operation of a market. It listed that company on the market it regulates. It therefore writes and enforces the Listings Requirements that it must itself comply with as an issuer. Every other JSE-listed company is supervised by an entity that is also its peer, its counterparty, and — in the case of the banks that own stakes in the market's settlement infrastructure — its business partner.

Global exchanges have almost all done this, so JSE is not an outlier. But the self-referential structure is not a curiosity. It is the root of a governance thread that runs through this entire story: JSE's authority rests on the market believing it is a neutral, competent gatekeeper. When that belief is tested — by a fraud it failed to catch, by a transparency refusal, by an antitrust allegation that its rules favour itself — the challenge lands on the same institution that is simultaneously trying to sell listings and grow earnings. Those two jobs are not always aligned, and Section V is where the tension becomes a legal proceeding rather than a philosophical one.

For now, the takeaway is simpler. Everything from 2005 onward is the story of a for-profit market-infrastructure company with a duty to shareholders. Which means it needed to start behaving like one — buying things, integrating them, and extracting more revenue per unit of market activity than a members' club ever would have dared.

↑ Back to Top

III. Building the Modern Machine: SAFEX, BESA, Strate, and the Logic of Vertical Integration (2001–2022)

If you want to understand what JSE Limited actually is today, stop thinking about the trading floor and start thinking about a pipeline.

A share trade in South Africa passes through a sequence of steps, each of which someone charges for. An order is matched on a trading venue. The resulting obligation is novated and risk-managed by a clearing house. The securities and cash change hands at a central securities depository. The broker's own books are reconciled and its clients' positions maintained in a back-office system. Data about the whole process is packaged and sold. Indices built on the prices are licensed.

Over roughly two decades, JSE quietly bought or built a position in almost every one of those steps. That is not diversification. It is vertical integration — the deliberate accumulation of tolls along a single pipe. It is also the reason the company's profits have proven far more durable than the number of listed companies would suggest.

2001: SAFEX, and the discovery that derivatives are better than equities

The first move was the South African Futures Exchange. JSE acquired SAFEX in 2001, absorbing both its financial derivatives and its agricultural commodity derivatives markets. At the time this looked like a modest bolt-on. In retrospect it was the most strategically important acquisition in the company's history, for a reason that has nothing to do with size.

Derivatives revenue behaves differently from cash equity revenue. Cash equity trading fees rise and fall with sentiment and value traded. Derivatives generate fees on volume and generate margin balances that sit at the clearing house — balances the clearer earns interest on and which grow with volatility rather than with optimism. In other words, derivatives are partially counter-cyclical to the thing that makes equity revenue fragile. When markets get frightening, cash trading can dry up while hedging activity and margin requirements spike.

SAFEX also gave JSE the agricultural franchise — the white maize, yellow maize, wheat and soya contracts that South African farmers, millers and traders use to hedge. That business has cultural weight in the country far beyond its revenue line, and it later became the site of an instructive failure, which we will come to.

2009: BESA, and a documentation gap worth admitting

In 2009 JSE bought the Bond Exchange of South Africa. The Competition Tribunal approved the acquisition unconditionally on 3 June 2009, and BESA became a wholly-owned subsidiary on 22 June 2009 through a scheme of arrangement.⁵ With it, JSE gained the country's fixed-income listing and trading infrastructure, folding bonds into the same house as equities and derivatives.

The strategic logic is straightforward. South Africa has an unusually large and liquid government bond market relative to the size of its economy, and it is the collateral backbone of the entire domestic financial system. Owning the fixed-income venue meant owning a piece of an activity that persists in every market condition — arguably more reliably than equity trading, since sovereign debt gets issued and traded whether or not anyone wants to buy shares.

Here an honest gap has to be flagged rather than papered over. The consideration JSE paid for BESA is not recoverable from the sources reviewed for this piece. The Tribunal record confirms the transaction and its approval but the merger documents in the public domain do not yield a clean purchase price, and JSE's contemporaneous disclosure of the deal economics is not readily accessible seventeen years later.⁵ That matters, because it means the single most useful capital-allocation test available for that era — did JSE overpay for scarce fixed-income infrastructure, or steal it? — cannot be run with confidence. Anyone who tells you the BESA deal was brilliant capital allocation is asserting something the public record does not currently support. What can be said is narrower and still useful: the fixed-income and financial derivatives franchise that came out of it now contributes a meaningful, growing revenue line, which is consistent with the asset having compounded, and inconsistent with a write-off. That is the honest ceiling on the claim.

Strate: the quiet royalty nobody talks about

The most under-discussed asset on JSE's balance sheet is not an operating business at all. It is a 44.55% equity-accounted stake in Strate, South Africa's central securities depository — the entity that actually holds and moves the country's securities, alongside historical shareholdings held by the major South African banks.⁶

Explain what a CSD does in plain terms: it is the country's single securities ledger. Every share, bond and money-market instrument that settles in South Africa settles across Strate's books. There is one of them. There is regulatory and practical room for exactly one of them. Every trade on the JSE settles there — and, crucially, so does every trade executed on a competing venue, because the challenger exchanges do not have their own depositories.

That is as close to a genuine royalty as exists in financial market infrastructure: capital-light, essentially unavoidable, and indifferent to where a trade was matched. In FY2025, JSE's share of Strate's economics contributed roughly R192 million, growing 20.7% — one of the fastest-growing lines anywhere in the group, delivered with almost no incremental capital.⁷

There is a subtlety worth holding onto for Section V. Because Strate settles trades from rival venues too, a share shift from JSE to a competitor does not destroy the Strate royalty. It only destroys the trading and clearing fee. That is a real, structural cushion under JSE's competitive risk — and it is the strongest single piece of evidence for the "toll collector" framing of this company.

JSE Clear (2022): making a licence into a line item

In 2022 JSE's clearing function became an independently licensed clearing house — JSE Clear — under South Africa's post-crisis regulatory architecture. This was primarily a compliance-driven restructuring, but it had a commercial consequence: clearing became a separately reported, separately priced business rather than a cost centre buried inside derivatives.

The results have been good. JSE Clear generated R130 million of revenue in FY2025, up 9.8%, and then accelerated sharply, growing 23.3% to R142 million in the first half of 2026.⁷⁸ That acceleration is not a mystery: clearing revenue rises with derivative volumes and with the margin balances the clearer holds, both of which were elevated through a very active market. It is also, therefore, cyclical — a point worth remembering when this line is offered as evidence of structural diversification.

The technology decision: renting the engine

The last piece is the least glamorous and possibly the most consequential. In 2012 JSE migrated its core cash equities trading onto the Millennium Exchange platform licensed from London Stock Exchange Group, upgrading it again in 2019. The physical trading infrastructure was also relocated so that South African order flow no longer had to round-trip to London — a latency improvement that made co-location a sellable product.

The strategic read is mixed and should be held as mixed. Licensing a world-class matching engine from a larger peer let a mid-sized emerging-market exchange run institutional-grade technology it could never have built alone. It also means JSE's most critical system is somebody else's intellectual property, and that JSE's competitive advantage in trading technology is, by construction, rented rather than owned.

That matters more when you notice what JSE did not migrate. Its back-office system — Broker Dealer Accounting, or BDA, which maintains broker books and client positions for the entire South African broking community — remained on legacy infrastructure that by 2021 was reportedly around thirty-five years old.⁹ On 18 August 2021, after a record R145 billion trading day overwhelmed it, BDA failed and the JSE could not open for five and a half hours.⁹¹⁰

Hold that fact. BDA is about to become the most important three letters in this story — the system at the centre of an operational failure, a modernisation capex programme, and a live antitrust case, all at once.

The net verdict on two decades of JSE M&A: this was consolidation of adjacent market infrastructure, not empire-building. No unrelated diversification, no overseas adventures, no obvious value destruction. It is a coherent pattern. Whether it was good capital allocation is a question the public record only partially answers, because the price paid for the largest deal is not recoverable. What can be observed is that the assets acquired are still there, still growing, and still contributing — which is more than can be said for a great many exchange acquisitions of the same era.

Which brings us to what all of that plumbing actually earns.

↑ Back to Top

IV. How JSE Actually Makes Money Today (Segment Deep Dive)

Imagine standing in the JSE's Sandton building on a heavy trading day in early 2026. Nothing dramatic is visible. There is no floor, no shouting, no paper. Somewhere in a data centre, a matching engine is processing an order book, and every few microseconds a fraction of a basis point is falling into the company's revenue line. Repeat, for eight and a half hours, for two hundred and fifty days a year, and you have most of a business.

But only most. The interesting thing about JSE Limited in 2026 is how much of its income no longer comes from that.

The group picture

For the year ended 31 December 2025, JSE reported revenue of R3,401 million, up 14.4%, and operating income of R3,535 million, up 14.2% — the gap between the two being non-operating and interest-related income the group counts in its topline measure.⁷¹¹ Net profit rose 16.7% to R1,071 million and headline earnings per share rose 17.7%.¹ Momentum carried into 2026: in the six months to 30 June, operating income grew 14.6%, net profit 16.9%, and headline earnings per share 18.8% to 816.2 cents, with the EBITDA margin expanding 100 basis points to 43.1% and operating cash flow up 20.6% to R624.7 million.⁸

Two observations before the segments. First, this is a business converting revenue growth into faster earnings growth — operating leverage is real here, because the incremental cost of processing another billion rand of trading is close to nothing. Second, a 43% EBITDA margin is healthy but not extraordinary by exchange standards; the largest global operators run materially higher. JSE is a good business, not a spectacular one, and part of the reason is that it carries regulatory, surveillance and issuer-supervision costs that a pure trading venue does not.

Capital Markets: the engine, and the cycle

Capital Markets is the largest segment and the one that made 2025 look like a breakout year. It generated R1,286 million in FY2025, up 17.8%, roughly 38% of group revenue.⁷

Inside it, the standout was equity trading: R571 million, up 28.5% — the fastest-growing major line in the entire company.⁷ That growth had almost nothing to do with JSE winning new business and almost everything to do with a surge in the value of shares changing hands on a market that was suddenly interesting to foreigners again. Management has consistently attributed the 2024–2026 trading surge substantially to the formation of the Government of National Unity following South Africa's May 2024 election, and the re-rating of South African risk that followed it.¹²

The rest of the segment tells you how diversified the trading franchise has become: primary market and listings fees of R194 million, equity derivatives of R130 million, bond and financial derivatives of R156 million, commodity derivatives of R94 million, and colocation fees — renting rack space next to the matching engine to firms who want their orders to arrive first — of R54 million.⁷

Note the proportions, because they overturn a common assumption. Listings fees, the revenue line most directly tied to the number of companies on the board, are roughly R194 million against a R3.4 billion group. Listings are about six percent of revenue. This is the single most important structural fact about the delistings debate, and it cuts in JSE's favour: the exodus of companies does not hit JSE where it earns. It hits JSE where it earns later, by shrinking the pool of things to trade.

In the first half of 2026, Capital Markets grew 17.6% to R719 million, with average daily value traded up 22.5% to R32.6 billion.⁸ And here is where management did something genuinely useful. On the H1 2026 call, they told investors that value traded was expected to "soften further and to normalize in H2."⁸ That guidance belongs immediately next to the growth figure, not in a risk appendix. The company's fastest-growing revenue line is one its own management has said will decelerate — which means anyone extrapolating 2025–26 equity trading growth into a durable trend is extrapolating against the operator's own view.

Post-Trade Services: the second pillar, and the legal exposure

Post-Trade Services earned R1,082 million in FY2025, up 17.5%.⁷ Its largest component was clearing and settlement at R548 million, growing an extraordinary 33.8%, followed by back-office BDA services at R432 million and funds under management at R102 million.⁷ In the first half of 2026 the segment grew 16.5% to R619 million.⁸

This is, in principle, the highest-quality revenue in the group. Post-trade is what happens after the decision to trade — it is closer to a utility than a market. Brokers cannot opt out of settlement. They cannot easily opt out of BDA either, because their entire client-account infrastructure runs on it and rebuilding that is a multi-year project no mid-sized South African broker wants to fund.

That last sentence is either a description of a magnificent switching-cost moat or a description of an antitrust problem, depending on who is reading it. As of 2026, a competition regulator is reading it the second way.

Information Services: structural growth, currency friction

Information Services — market data and FTSE/JSE index licensing — earned R498 million in FY2025, up 9.8%, and R273 million in H1 2026, up 7.3%.⁷⁸

This is the segment that most resembles the businesses global investors pay high multiples for. Data and index revenue is subscription-like, high-margin, and grows with the number of institutions and applications consuming it rather than with trading volumes. It is the closest thing JSE has to the economics that made the data divisions of LSEG and ICE so valuable.

It also carries a specific, quantified vulnerability. Roughly 61% of the segment's revenue is denominated in US dollars.⁷ Management disclosed that underlying dollar growth in H1 2026 was around 10%, compressed to 7.3% in reported rand terms purely by translation as the rand strengthened.⁸ That is not a hypothetical currency risk paragraph; it is a real, measured 270-basis-point haircut on the group's most attractive segment, in a single half. A stronger rand — which is generally good news for South Africa — is mechanically bad news for JSE's best business. Investors should hold that inversion in mind.

JSE Investor Services: the one going backwards

Not everything is growing. JSE Investor Services — the transfer-secretarial and share-plan administration business, whose economics depend substantially on interest earned on client cash balances — fell 7.1% to R212 million in FY2025 and declined a further 5.6% in H1 2026 to R102 million.⁷⁸

The mechanism is simple and unarguable: as South African interest rates fall, the margin earned on float shrinks. There is no strategic failure here and no management spin available; it is arithmetic. It is worth stating plainly for exactly that reason. A company where every segment always grows is usually a company with a presentation problem. JSE has a segment that shrinks when rates fall, discloses it, and explains why. That is a small but real credibility marker.

The diversification claim, and how much of it survives scrutiny

The strategic story management has told for several years is that a rising share of non-trading income insulates JSE from both the listings decline and trading cyclicality. The numbers attached: non-trading income represented roughly 38% of operating income in FY2024 and around 34% of revenue in H1 2026.¹²⁸

Is the claim true? Partly, and it is worth being precise about which part.

It is true that JSE now earns roughly a third of its income from sources that do not require anyone to press a buy button today — data, index licensing, listings fees, investor services, custody-adjacent revenue. That is a genuine structural improvement over the pre-2010 exchange, and it is the strongest evidence for the "infrastructure toll" framing.

But the claim is weaker than it first sounds, for two reasons the company's own disclosure supports. First, several lines counted outside "trading" are still activity-sensitive: clearing revenue scales with derivative volumes and margin balances, and market data consumption scales with how many people are actively trading South African assets. Second, the two most reliably non-cyclical lines — index licensing and listings fees — are precisely the ones exposed to the currency and the listings pool respectively. The diversification is real. The insulation is partial. The honest formulation is that JSE has meaningfully reduced its beta to South African equity turnover, not eliminated it.

JSE Private Placements: optionality, sized honestly

One initiative deserves a paragraph and no more than a paragraph, because its economics do not justify more.

JSE launched a private placements platform in 2021 and expanded it in October 2024 through a partnership with UK fintech Globacap, pitching it publicly as the exchange's answer to the private capital that has been absorbing companies which would once have listed. The strategic framing is sound. The commercial reality is that the platform generated R327,000 in fees in 2025 — down from R831,000 in 2024.⁷

Three hundred and twenty-seven thousand rand, against a R3.4 billion group, is a rounding error that shrank. This is real optionality on private-markets share and it is not yet a business. The reason to mention it at all is not the number; it is the pattern. Section VII returns to this, because a company that has repeatedly announced strategically-framed initiatives with negligible revenue attached has, by that record, earned a degree of scepticism about the next set of strategically-framed initiatives.

For now, the picture is a company earning good money from a well-integrated pipeline, with one segment booming cyclically, one segment structurally attractive but currency-exposed, one segment declining for reasons of arithmetic, and one segment — the second largest — sitting directly in the path of a regulator.

That regulator is where the story gets serious.

↑ Back to Top

V. The Moat on Trial: Industry Structure, A2X, and the Competition Tribunal

On 10 November 2025, the South African market learned that the Competition Commission had, on 1 October, referred JSE Limited to the Competition Tribunal for abuse of dominance.¹³¹⁴ The Commission was seeking a penalty of up to 10% of the exchange's annual turnover, plus an order compelling JSE to change its rules so that inter-exchange trading became easier — including allowing JSE trades to be executed, cleared and settled on a rival venue.¹³

JSE denied the allegations "in the strongest possible terms," said its external counsel considered the claims without merit, and indicated its formal plea would be filed in early 2026.¹⁵¹⁶ As of this writing, the matter remains unresolved.

This is not a footnote risk. It is the most important open question about the business, because it goes directly at the mechanism that makes JSE valuable.

What an exchange moat actually is

Before the case, the theory. Exchanges are one of the cleanest textbook examples of a network-effects business, and it is worth being precise about why, because the precision is what the Tribunal case attacks.

Liquidity is self-reinforcing. A buyer goes where the sellers are; a seller goes where the buyers are. The venue with the deepest order book offers the tightest spread, which attracts more flow, which deepens the book. This is a positive feedback loop that, once established, is brutally difficult to break — it is why challenger exchanges in almost every market have historically struggled, and why the incumbent's share is usually far more durable than its technology or its pricing would justify.

In Hamilton Helmer's 7 Powers vocabulary, an incumbent exchange typically holds three at once. Network economies, described above. Switching costs, because brokers must build and certify connectivity, integrate clearing and settlement, and retrain operations staff to route flow anywhere new. And scale economies, because the marginal cost of matching an additional trade is effectively zero, so the largest venue has structurally the lowest unit cost. Add cornered resource in JSE's case — the Strate stake, and the regulatory licences themselves — and you have as complete a moat structure as exists in financial services.

Run Porter's five forces and the picture is similar. Barriers to entry are extraordinarily high, both technically and through licensing. Supplier power is limited, though the LSEG technology dependency is a genuine, if modest, exception. Substitutes exist — companies can list abroad, investors can trade South African exposure via depositary receipts or derivatives offshore — and that substitution threat turns out to matter a great deal, as Section VI shows. Buyer power is fragmented among many brokers and asset managers, none individually essential.

The one force that has changed is rivalry. And the question the Tribunal will decide is whether JSE won that rivalry fairly.

A2X: the challenger that did not go away

A2X Markets was founded on 10 October 2014 by Sean Melnick, Ashley Mendelowitz and Kevin Brady, licensed by the regulator in 2017, and opened for trading on 6 October 2017 with three listings — African Rainbow Capital, Peregrine Holdings and Coronation Fund Managers — worth a combined R14 billion.¹⁷ African Rainbow Capital, Patrice Motsepe's investment vehicle, had taken a 20% stake in March 2017.¹⁸

The A2X pitch is not technological romance. It is price. A2X charges materially less than the incumbent to match a trade, and claims participants save around 40% on execution costs.¹⁹ Its model is secondary listings: a company keeps its primary listing on the JSE and adds a secondary listing on A2X, so the same share can trade on either venue and brokers can route to whichever is cheaper or deeper. No company has to leave the JSE. That is what makes A2X hard to dismiss — it asks issuers for almost nothing.

The roster it has assembled is not trivial. AngloGold Ashanti, Sanlam, Discovery, Standard Bank, Prosus and Naspers have all taken A2X secondary listings, and in September 2026 Capitec — one of the most heavily traded shares in South Africa — began trading on A2X from the 7th, while retaining its JSE primary listing.²⁰²¹ These are not marginal names. They are the index.

Here the analysis has to be careful, because the published numbers genuinely disagree, and the disagreement is not innocent — different parties are measuring different things.

A2X's own growth metrics are real: trade value grew from R657 million in 2017 to more than R8.33 billion in a single month by August 2024, across more than 180 securities with a combined market capitalisation of R9.42 trillion.¹⁹ But note what that last number is. R9.42 trillion is the market capitalisation of the companies eligible to trade on A2X, not the value A2X actually trades. It measures the size of the shop window, not the till.

The Competition Commission's complaint contains a different and much more striking figure: it puts JSE's share of secondary trading at 62% as of December 2024, against 38% for A2X — placing JSE above the 45% dominance threshold in South African competition law.²² Take that at face value and A2X is not a niche challenger; it is a genuine duopoly partner.

But that figure sits uncomfortably against every independent trading-value estimate, which has generally placed A2X's share of South African equity turnover in the low single digits — under 1% in its earliest years and in the region of 2–3% by the mid-2020s. Those two claims cannot both describe executed rand volume. The most plausible reconciliation is that the Commission's measure captures something narrower or differently defined — a subset of trade types, or a measure weighted by listed value rather than value traded. Without the underlying referral methodology, this cannot be settled here, and it should not be pretended otherwise.

What can be said with reasonable confidence is this. A2X has captured real and growing share in the specific, valuable niche of highly liquid blue-chip secondary trading. It has not achieved a structural shift in overall South African equity turnover after nearly nine years. And the aspirational targets its backers articulated around 2022 — 15% to 25% of the market within five years — have, on the trading-value measures, not been met.

That is the honest calibration, and it points in a direction that will annoy both camps: the "moat is eroding" claim is not rejected by the evidence, but it is narrowed considerably. A2X is a persistent margin threat in JSE's most profitable order flow, not an existential one. What would change that verdict is a share figure, on a consistent executed-value basis, breaking sustainably into double digits — that is the number to watch, not the listings count.

The case itself, and why it is the right test

The Commission's complaint, filed by A2X in October 2022 and investigated for three years, centres on two things.¹³²²

The first is BDA. The Commission alleges JSE has effectively mandated the use of its Broker Dealer Accounting system, and that BDA is not compatible with A2X's systems, creating a barrier to entry.²² Translate this into operational reality. A South African broker's client positions, statements and reconciliations live in BDA. If a trade is executed on A2X, that trade still has to end up correctly recorded in BDA for the broker's books to make sense. If the interoperability between the two is poor, the broker faces extra manual work, extra reconciliation risk, and extra cost for every A2X trade — which is a powerful reason to simply route everything to the JSE regardless of price. In that scenario, JSE's execution share would be protected not by having the better market, but by controlling the plumbing every broker must use afterwards.

The second is the "matched principal" trade type — rules governing how brokers handle trades where they stand between two clients — which the Commission alleges are applied asymmetrically in a way that makes routing to A2X costlier.

The reason this belongs in the middle of the moat discussion rather than in a risk list is that it is the cleanest available test of the distinction that matters. Network effects are an earned advantage: customers stay because staying is better. Control of mandatory infrastructure is an enforced advantage: customers stay because leaving is artificially expensive. Both produce high market share and high margins. They have very different durability and very different regulatory futures.

Three outcomes are worth thinking through. If the Tribunal dismisses the case, JSE's moat is validated in the strongest possible way — a competition authority looked hard for three years and found the dominance legitimate. If JSE loses on the conduct but the remedy is confined to interoperability rule changes, the direct financial hit may be modest but the strategic consequence is significant: the friction protecting JSE's most valuable order flow is removed by law, and A2X gets to compete purely on price in a business where JSE has more to lose. If JSE loses and is fined near the statutory maximum, the penalty could approach a full year of profit — a serious, though survivable, capital event for a company with a strong balance sheet and a 22% return on equity.

There is a second-order point worth noting. The Commission's proposed remedy includes allowing JSE trades to be cleared and settled on A2X.¹³ That is a more radical ask than a rule tweak, and it goes at the post-trade segment — the R1,082 million business — rather than just at execution fees. The economic exposure here is larger than a headline reading of "trading fees" would suggest.

And there is an unmistakable irony in the fact that the system at the centre of the allegation is the same thirty-five-year-old system that took the market down for five and a half hours in 2021, and the same system now absorbing the largest share of JSE's FY2026 capital expenditure.⁹⁸ The legacy technology is simultaneously an operational risk, a competitive weapon, and a modernisation cost.

ZAR X: the cautionary tale that cuts both ways

There was a third exchange. ZAR X launched in 2017 with a genuinely differentiated pitch — real-time, T+0 settlement, versus the multi-day cycle everyone else ran — aimed particularly at restricted-share and empowerment-scheme trading.

It did not die of product failure. The FSCA suspended its exchange licence on 20 August 2021 for non-compliance with the liquidity and capital adequacy requirements of section 8 of the Financial Markets Act, and cancelled it outright on 13 February 2023 after ZAR X proved unable to meet those requirements during the suspension.[^23]²³ South African exchanges must hold capital equivalent to at least six months of operating expenses.[^23] The Public Investment Corporation held a 24.14% stake, and an unresolved capital-raise dispute sat behind the collapse.[^23]

The correct reading holds two things simultaneously, and the outline is right to insist they not be collapsed into one.

On one hand, this is strong evidence for JSE's structural durability. Regulatory capital adequacy is a real, non-negotiable barrier to entry. A challenger cannot bootstrap its way into the South African exchange business on venture-scale funding and a good idea; it needs a permanent balance sheet, which means it needs patient institutional capital before it has any revenue. Most challengers will fail this test, and ZAR X did.

On the other hand, A2X — over exactly the same period, with African Rainbow Capital and Nala Empowerment behind it — passed it, and is still there nine years in, growing, and now litigating. The barrier is high. It is not insurmountable. Complacency about it is unwarranted.

The niche competitors round out the picture: 4AX, now the Cape Town Stock Exchange, serves small and mid-cap and debt issuers with a combined listed market capitalisation in the region of R6.7 billion — roughly a rounding error against the JSE's multi-trillion-rand board, and not a competitive factor in the large-cap trading business that generates JSE's revenue.

Governance as competitive asset — and where it has frayed

There is one more dimension of the moat that does not appear in any framework, and it is the one that would be most expensive to lose: trust. An exchange's franchise ultimately rests on issuers, investors and regulators believing it enforces its own rules competently. Which makes JSE's record as a gatekeeper an investment question rather than an ethics question.

That record is mixed, and it is worth being precise about how mixed.

Two of the largest corporate frauds in South African history — Steinhoff, exposed in December 2017, and Tongaat Hulett, exposed in 2019 — involved JSE-listed companies whose misstatements ran undetected for years.²⁴ A nuance matters here and is frequently lost: Steinhoff's primary listing was in Frankfurt, with the JSE holding a secondary listing, which limited JSE's primary supervisory role. Tongaat was squarely JSE's. In 2020, JSE publicly censured Tongaat and imposed a fine of R7.5 million — the maximum available to it — for financial statements from 2011 to 2018 that were "incorrect, false and misleading" in material respects, with R2.5 million of that suspended for five years.²⁵²⁶

Seven and a half million rand, for eight years of falsified accounts at a company that destroyed billions in shareholder value. That is not a failure of will; it is a statutory ceiling. JSE's own issuer-regulation director confirmed in 2026 that the exchange simply cannot impose penalties above R7.5 million under the Financial Markets Act.²⁷ It is a structural limit on JSE's regulatory teeth, and it is worth understanding that JSE's gatekeeper role is therefore substantially about disclosure enforcement and censure rather than deterrence-grade punishment. The FSCA, which fined Tongaat R20 million separately, carries the heavier weapons.²⁵

The volume of enforcement work is not trivial. In its 2025 financial year, JSE's issuer regulation division handled 150 investigations — 43 carried over and 107 new — completing 106 by year-end with 44 ongoing, covering late material disclosures, unauthorised transactions, director qualification misrepresentation and directors' dealings breaches.²⁷ That is an active supervisor, working within a small statutory penalty box.

Against that, a specific and sourced criticism. In August 2025, amaBhungane documented JSE's refusal of a Promotion of Access to Information Act request from Inhlanhla Ventures, a minority shareholder investigating suspected manipulation in enX Group shares — the shares had collapsed from 700c to 320c in mid-May 2020 and then recovered — seeking the identities of buyers and sellers and the sale values.²⁸ JSE cited confidentiality and the Financial Markets Act. South Africa's Information Regulator found the refusal unjustified; JSE threatened to take the finding on review, and enforcement remained pending.²⁸

How much weight should this carry? Less than a pattern, more than nothing. It is one named episode, bounded to a specific request and a specific company, and it does not establish systemic capture. But it goes to a real question about a self-regulating, self-listed market operator: when transparency is inconvenient, which instinct wins? On this occasion, on the Information Regulator's own finding, the answer was secrecy — and JSE fought the finding rather than accepting it.

The conclusion the evidence supports is narrower than either extreme. JSE is neither an unimpeachable gatekeeper nor a captured one. It is an active, statutorily under-armed regulator with a documented instinct toward institutional confidentiality, operating with an inherent conflict between selling listings and policing them. For an investor, that translates into a specific vulnerability: the franchise depends on trust, trust is periodically tested, and JSE has so far survived those tests with reputational damage but no loss of licence or mandate.

Which leads to the harder problem — the one that has nothing to do with competitors or regulators, and everything to do with there being fewer and fewer companies to regulate at all.

↑ Back to Top

VI. The Listings Exodus: South Africa's Shrinking Public Market

In March 2026, the man responsible for conduct regulation across South Africa's entire financial sector stood up and said, in effect, that the country's public equity market was becoming too small to do its job.

Unathi Kamlana, Commissioner of the Financial Sector Conduct Authority, warned publicly that the decline in local listings raised questions about the depth and vibrancy of South Africa's public capital markets and their capacity to support economic growth.³²⁹ From roughly 850 listings in the mid-1990s, the exchange was down to about 280 companies, with more than 500 delistings over twenty-five years.³

Kamlana's framing was the sharp part. The delistings, he suggested, were the visible symptom. The deeper problem was that fewer companies were choosing to list in the first place — and that the firms exiting skewed smaller and less profitable, concentrating market capitalisation among a shrinking number of very large companies.³

That is a systemic concern for South Africa. Is it an existential concern for JSE Limited?

Myth versus reality: how bad is the shrinkage, actually?

The consensus narrative — that the JSE is dying — is stronger as rhetoric than as arithmetic, and it deserves to be fact-checked properly rather than repeated.

The most substantive counter-argument comes from Allan Gray, one of South Africa's largest and most respected asset managers, which has argued publicly that the "shrinking JSE" framing is overstated.³⁰ Its case has four planks.

First, the raw count overstates the loss of substance. The exchange went from 776 companies to 331 over roughly thirty years, an average of about fourteen delistings a year — but much of that reflects consolidation rather than exit.³⁰ The illustration is devastatingly good: in 1982 the mining sector alone had 93 listed companies, including 45 individual gold mines. Today roughly seven locally listed gold miners remain — each of which owns a portfolio of the mines that used to be listed separately.³⁰ The gold did not leave the JSE. It got aggregated into fewer, bigger holding companies. Counting listings treats that as a 38-company loss. Counting economic exposure treats it as roughly neutral.

Second, the market capitalisation of new listings has exceeded the market capitalisation of delistings every year since as far back as 2008.³⁰ Companies leaving have been predominantly smaller mid-caps; companies arriving have been larger. Since JSE's revenue is driven by value traded rather than by company count, this is precisely the mix shift that keeps revenue growing while the headline number falls.

Third, JSE-listed companies derive roughly 70% of their revenue internationally, making the exchange far less domestically dependent than most.³⁰ A South African index is, to a substantial degree, a claim on global earnings.

Fourth, the trend is not uniquely South African. Public-market listing counts have fallen across most developed markets for the same reasons — the growth of private capital, the regulatory and disclosure burden of being public, and the willingness of large private funds to keep companies out of public markets for far longer.

Against this, one uncomfortable data point resists the benign reading. "Vanishing Acts," an econometric study of South African firm delistings by the University of Cape Town's Development Policy Research Unit, co-authored by economics professor Haroon Bhorat and commissioned by the Association for Savings and Investment South Africa, found that the JSE's average annual delisting rate over 1993–2015 ran at 7.8% — almost double the 4.1% global mean.⁶⁰⁶¹ Worse for the bull case, new listings ran at 4% of the total against a global average of 6.8%, which is the arrival side of the ledger rather than the departure side.⁶⁰ The study's own mitigating finding is that roughly 80% of the exodus happened before 2005 and that most exits were merger-and-acquisition driven rather than distress-driven — which supports Allan Gray's consolidation reading.⁶⁰ But a market that simultaneously loses companies faster than the world and gains them slower than the world is not merely consolidating. It is under-replacing.

The calibrated verdict: the strong bear claim — that the JSE is structurally dying and the count decline maps to economic decline — is not supported. The listing count materially overstates the damage, and the revenue evidence proves it, since JSE has grown earnings through the entire period of decline. But the weaker bear claim survives intact: a market with fewer, larger, more internationally-exposed companies is a market with less domestic breadth, fewer new entrants, and a narrower base from which future large companies can emerge. It is a slower-burning problem than the headlines suggest, and it is still a real one.

The flow that actually matters

Annual listing flow is where the abstraction becomes measurable, and the recent record is genuinely poor before it is genuinely better.

2023 was the trough. Just three new companies came to market — Premier Foods, Copper360 and Primary Health Properties — against 24 JSE delistings, plus three more companies displaced when ZAR X closed.³¹ A ratio of one arrival for every eight departures is not a cycle; it is a structural signal, and it was correctly read as one.

2024 improved, but against management's own guidance it merely landed rather than beat. Leila Fourie had publicly anticipated as many as ten new listings for the year.³² By mid-October the count stood at five, with the debut of Altvest Capital on AltX.³³ The year closed at eight new listings.³⁴ Eight against guidance of "up to ten" is a soft outcome — not a miss, not a beat, and a useful early data point on how to read this management team's numeric targets: directionally honest, mildly optimistic.

2025 is where the picture becomes genuinely more interesting, because the count and the value diverged sharply. The headline number of new listings was not obviously better than 2024's.³⁴ But the quality was transformed. On 4 November 2025, Optasia — an AI-driven fintech processing more than 30 million loan transactions daily across 38 countries — listed on the Main Board at R19 per share, raising gross proceeds of approximately R6.5 billion at a valuation of roughly R23.5 billion.³⁵³⁶ It was the largest IPO in Africa that year and the largest fintech IPO in the wider EMEA region since 2021, and it was several times oversubscribed.³⁵

Three weeks later, on 27 November, Cell C listed after raising R2.7 billion at R26.50 per share through the sale of 102 million shares by The Prepaid Company, giving the mobile operator an indicative market capitalisation of about R9 billion.³⁷³⁸ The debut was muted — the shares opened flat — which is its own signal about depth of domestic demand for a large new issue.³⁹

The read: South Africa's IPO market in 2025 was not broad, but it was capable of absorbing large, high-quality issues at scale when they came. That is a materially different diagnosis from "the market is closed." A market that can price a R23.5 billion fintech several times oversubscribed has functioning institutional demand. What it lacks is a pipeline of mid-sized companies choosing to use it.

The structural case studies: why companies leave, and one that didn't

The abstraction becomes concrete in a handful of departures that shaped the narrative.

Anglo American moved its primary listing to London in 1999, keeping only a JSE secondary listing, explicitly to access deeper pools of capital — the original template for the flight to London, and the one that established in South African corporate minds that leaving was both possible and rewarded. Naspers followed the same logic in 2019, taking its Prosus vehicle to Euronext Amsterdam, driven by a different problem: Naspers had grown so large relative to the JSE that South African institutional investors were structurally unable to hold enough of it, creating a persistent discount that only an offshore listing could address. South32, demerged from BHP in 2015, chose an ASX primary listing with JSE secondary. Richemont, meanwhile, is often miscounted in this narrative — it modernised its JSE access in 2023 by converting from a depositary-receipt structure to a direct secondary listing, but it was never JSE-primary, and its 2023 change was an improvement in South African investors' access, not a departure.

Set against those, two counter-examples matter more than they are usually given credit for.

Bidcorp's 2016 unbundling from Bidvest created a large, successful, entirely domestic spin-off that stayed JSE-primary — proof that South African corporate structures can still generate new large-caps internally.

And Valterra Platinum, the most recent and most telling. When Anglo American demerged its platinum group metals business — a transaction that completed on 31 May 2025 following shareholder approval on 30 April — the new company kept its primary listing on the JSE and took only an international secondary listing in London, where it began trading on 2 June 2025 under the ticker VALT.⁴⁰⁴¹ A business of that scale, freshly separated from a London-headquartered parent, with every incentive and every adviser pointing to London, chose Johannesburg as its home market.

That single decision cuts hard against a purely one-directional narrative. Companies with deep South African operating assets, South African shareholder registers and South African index inclusion still have real reasons to be JSE-primary. The flight to London is a pattern, not a law.

The reversal that is actually new

The most material development for the listings debate is not a South African company staying. It is foreign companies arriving.

In October 2025, Coca-Cola HBC agreed to acquire a 75% controlling interest in Coca-Cola Beverages Africa from The Coca-Cola Company and Gutsche Family Investments for approximately $2.6 billion — around R45 billion — and announced that it would pursue a secondary listing on the JSE upon completion, to reinforce its long-term commitment to the continent.⁴²⁴³ By July 2026 the deal had cleared a key regulatory hurdle, moving the listing closer.⁴⁴ It is expected to be the largest listing on the JSE in a decade, entering the Top 40 around the fifteenth position.⁴⁴

Separately, Canal+ committed to an inward secondary JSE listing following its takeover of MultiChoice — a transaction valued at roughly R55 billion, after which MultiChoice itself delisted from both the JSE and A2X in December 2025. As of mid-2026, JSE management stated that both the Coca-Cola HBC and Canal+ secondary listings remained on track for 2026.⁴⁵

Two things should be said about this, and they pull in opposite directions.

The first is that these are real, large, and genuinely new. They reverse a two-decade assumption that capital flows out of Johannesburg and never back. Both arrive as a direct consequence of foreign companies acquiring large South African operating businesses and wanting local shareholder participation — a structural channel that did not previously exist at scale.

The second is that they are secondary listings won because of M&A, not primary listings won because South African companies chose to be public. They add tradeable value and index weight, which is exactly what JSE's revenue model likes. They do not address Kamlana's actual concern, which is the absence of new domestic issuers. Management's framing that the "crisis is easing" is management's framing, dated to 2026 reporting, and should be read as such rather than adopted.⁴⁶

The policy response, tracked over time

The most useful test of a management team is not what it says at one point in time but whether its story moves coherently as facts change. On listings, JSE's arc is trackable and, on balance, holds up.

In 2023 and early 2024, the tone was quiet optimism with a specific number attached — up to ten new listings in 2024.³² The number was roughly met, not clearly beaten.

By 2025 the language hardened. Fourie publicly acknowledged an IPO crisis rather than a soft patch — a notable choice for the chief executive whose revenue depends on the market not being in one.⁴⁶ That candour was paired with concrete action rather than commentary. The FSCA approved JSE's Simplification Project, which rewrote the Listings Requirements in plain language and cut their volume by more than half, and approved the Market Segmentation Project, which split the Main Board into a Prime Segment and a General Segment effective 23 September 2024, offering meaningfully lighter regulation to smaller issuers outside the All Share Index while preserving disclosure standards.⁴⁷⁴⁸ The fast-track framework for secondary listings from recognised foreign exchanges was extended to cover additional venues including Euronext, Tadawul and Hong Kong.

Beyond its own rulebook, JSE convened Operation Phumelela — Zulu for "to succeed" — a financial-sector competitiveness taskforce launched in 2024 in partnership with National Treasury, the South African Reserve Bank and the FSCA, chaired by Fourie.⁴⁹ Its most visible win came on 24 October 2025, when the FATF plenary in Paris removed South Africa from the grey list after 32 months of enhanced monitoring, following implementation of all 22 required action items.⁵⁰⁴⁹ Grey-listing had raised transaction costs and compliance friction for every cross-border flow into the country; exiting it removed a concrete deterrent to foreign participation.

That is a substantive record: candour about the problem, rule reform delivered and approved, and a policy coalition that produced a measurable national outcome. It is also, so far, a record of inputs rather than outputs. When Fourie's retirement was announced, a Business Day editorial published on 15 October 2025 assessed her as "a moderniser, not a redeemer" — crediting the modernisation, the improvement in earnings quality and the expanded strategic footprint, while concluding that her reforms had made the JSE more defensible rather than more magnetic.⁶² That distinction is the right one, and it is the crux of the listings debate. Defensible protects the earnings JSE already has. Magnetic would grow the pool. Only the first has been demonstrated.

So the claim "JSE has solved its listings problem" is rejected by the evidence. The claim "JSE has taken credible, concrete action on its listings problem and the environment has improved" survives. The falsifying test is specific and available: whether net listings flow, weighted by market capitalisation, turns positive on a sustained basis across 2027 and 2028 — after the one-off Coca-Cola HBC and Canal+ arrivals have washed through the numbers.

Which raises the obvious question: who is now accountable for delivering that?

↑ Back to Top

VII. Leadership in Transition: Fourie's Record and Reddy's First Test

There is a particular kind of corporate moment that is difficult to analyse and important to get right: the handover from a successful long-tenured chief executive to an internal successor, at the exact point where the outgoing leader's numbers look their best.

JSE reached that moment in the first quarter of 2026. Leila Fourie announced her retirement in October 2025, delivered a record set of full-year results in March 2026, and left on the 31st.⁵¹¹ Valdene Reddy took over on 1 April.⁵¹ Five months later, in August, Reddy presented her first results and, alongside them, an entirely new five-year strategy.

Leila Fourie: what the record actually shows

Fourie was not an outsider parachuted in. She had already served as JSE's Executive Director for Post-Trade and Information from 2012 to 2016 before leaving, and returned as Group CEO in October 2019 — meaning she took the top job with direct operating experience of the two segments that would later become the group's diversification story. Her earlier career spanned Standard Bank's card division and the Commonwealth Bank of Australia, and she holds a doctorate.

The outcomes over her tenure are measurable and, on the whole, good. Return on equity rose from around 17% in 2019 to 22.0% in FY2025.¹ The combined market capitalisation of JSE-listed companies roughly doubled during her tenure, from about R12.6 trillion to over R24 trillion.⁵² Non-trading income grew from a minority contributor to roughly a third of operating income — the clearest single piece of evidence that the diversification strategy was executed rather than merely announced.¹² The board and press credited her with a turnaround in earnings quality, a diversified revenue base, and modernised technology and regulatory frameworks.⁵¹

She also left cleanly. There was no scandal, no activist campaign, no board dispute, and no pre-emptive succession announcement forced by underperformance. Her stated plan was to sail — she had, by her own account, an ambition to sail around the world, alongside rock climbing.² JSE Chairperson Phuthuma Nhleko's public comments framed the transition as planned rather than reactive.⁵¹

That is genuinely unusual and worth crediting. But credibility assessments should be made on the hard cases, not the easy ones, so consider the two places where the record is less flattering.

The first is listings. She named the problem honestly — using the language of a crisis rather than a soft patch, which is not what a chief executive under pressure usually does — reformed the rulebook meaningfully, and did not fix the underlying issue. The independent press verdict on that record was neither hostile nor forgiving, and it is the fair one.⁶² Candour about an unsolved problem is a genuine credibility marker. It is not the same thing as having solved it.

The second is conversion of strategic initiatives into revenue. Private Placements, launched in 2021 and expanded in 2024, was generating R327,000 a year by 2025 and shrinking.⁷ And in September 2025 the JSE suspended its basis futures contracts for grain — instruments that let traders hedge price risk at specific silo locations against the Randfontein benchmark — after four years of industry effort to establish them. The reason was partly technical: JSE's software could accommodate only a limited number of silos, and expanding beyond the initial ten pilot sites would have required new infrastructure, while the legal division raised fairness concerns about restricting the pilot.⁵³ The South African Grain and Oilseeds Trade Association's executive director called the decision regrettable and warned that "the JSE is in danger of losing its leading role," noting the industry would pursue alternatives including a competing service from Match Exchange.⁵³

That is a small revenue line and a big signal. A four-year product effort in JSE's own historic franchise — agricultural derivatives, inherited from SAFEX — was abandoned because the incumbent's technology could not scale it and its legal function could not resolve the design, with a competitor waiting to take the business. It is a concrete instance of legacy technology and internal process constraining commercial delivery. Anyone assessing JSE's ability to launch new products should weigh it.

Remuneration: pay, and what happened when shareholders objected

Fourie's total compensation in her final full year was approximately R26.1 million, split roughly 30% fixed and 70% variable, with a personal shareholding of around 0.7% of the company.⁵⁴ Those figures are unremarkable by global exchange-CEO standards and modest relative to the value created — worth stating plainly rather than editorialising in either direction.

The more revealing story is what happened when shareholders pushed back.

At the May 2024 AGM, JSE's remuneration policy received 78.03% support and its implementation report 78.31% — meaning roughly 22% of votes cast opposed each, uncomfortably close to the 25% threshold that triggers mandatory engagement obligations under South African governance practice.⁵⁵ For a company that writes and enforces the governance rules other issuers must follow, a 22% pay revolt is a genuinely awkward result.

What the board did next is the part that matters analytically. It engaged its top twenty shareholders, representing 78.4% of issued capital, and made specific structural changes rather than cosmetic ones: it lifted the weighting of hard financial metrics in short-term incentives to 60%, introduced bonus deferral, and — most tellingly — raised the return-on-equity stretch target for long-term incentive vesting from 19.5% to 23%.⁵⁶⁵⁴ Long-term incentive awards were set at 200% of guaranteed pay for the CEO, having been raised in a prior cycle to 250% specifically in response to shareholder pressure for more pay tied to long-term outcomes, with the CFO's LTI lifted from 140% to 160%, and minimum shareholding requirements imposed at 200% of pay for the CEO, phased over five years.⁵⁴

Support recovered to roughly 90% at the May 2025 AGM.⁵⁴

Raising your own ROE vesting hurdle by 350 basis points is not a public-relations gesture; it makes it harder for executives to be paid. This is a clean, documented instance of the governance feedback loop working: shareholders objected, the board engaged, incentives tightened, dissent fell. On the specific question of whether this board course-corrects under pressure, the evidence says yes.

Valdene Reddy: continuity, with everything still to prove

Reddy is an insider by design. She spent more than ten years at the JSE, most recently as Director of Capital Markets and previously heading Equity and Equity Derivatives, with over twenty years in financial markets and senior roles at international investment banks before that — Bank of America Merrill Lynch and Renaissance Capital.⁵¹ She holds a Bachelor of Business Science in Actuarial Science from the University of Cape Town, is a Certified Director with the Institute of Directors South Africa, and completed Harvard Business School's Advanced Management Program.⁵¹ She became the JSE's third woman CEO.

The relevant analytical point is not her CV but what her appointment signals. This was a continuity pick, not a reset. The board did not go outside for someone to change direction, which tells you it does not believe the direction needs changing. For investors, that means "new management risk" is a smaller variable here than it would be after an external hire — and equally that anyone hoping for a strategic discontinuity should not expect one.

FORGE 2031: the plan, and what would prove it

On 4 August 2026, alongside her first set of results, Reddy unveiled FORGE 2031 — a five-year strategy she described as "endorsed by the board and the roadmap to turn resilience into growth."⁸

It has two pillars. Transformation covers organisational redesign — already executed — technology modernisation, and cost discipline, aimed at margin expansion. Growth covers three things: a pan-African digital marketplace, commercialisation of data and services at higher margin, and monetisation of JSE's technology assets.⁸

The guidance attached is specific enough to be tested. Operating expense growth of 6–8% for FY2026 adjusted for one-off costs — against 11.5% reported in the half, or roughly 3.5% on an underlying like-for-like basis. Capital expenditure of R190–230 million, weighted toward BDA modernisation and a Bond Central Counterparty platform targeted for 2027. A dividend payout ratio target of 67–100%.⁸ And the deceleration guidance already noted: value traded expected to soften and normalise in the second half.⁸

Now assess it honestly. The transformation pillar is credible because it is mostly arithmetic — organisational redesign already done, cost growth guided, capex earmarked for a system that demonstrably needs replacing. The Bond Central Counterparty is the most interesting item on the capex list, because a bond CCP would extend JSE's clearing franchise into fixed income and generate the same margin-balance economics that make JSE Clear attractive. That is a coherent extension of the vertical-integration logic that has worked before.

The growth pillar is where scepticism is warranted, and the company's own record is the reason. "Pan-African digital marketplace" and "technology monetisation" are the kind of phrases that have appeared in exchange strategy documents for two decades and converted to revenue rarely. JSE has, within the last five years, launched a private placements platform that earns three hundred thousand rand a year and abandoned a four-year agricultural derivatives project because its systems could not scale it. That is a documented conversion record, and it should temper how much credit the market extends to a five-year growth narrative in advance of evidence.

The analyst response on the H1 2026 call was notably thin — a single substantive question on FORGE 2031 delivery, to which Reddy emphasised execution discipline, a phased investment approach, and AI deployment already underway, with broader shareholder engagement sessions promised over the following four to eight weeks.⁸ Thin Q&A is not necessarily a bad sign, but it means the strategy has not yet been publicly stress-tested by the sell side, and investors have less external scrutiny to lean on than they might assume.

The specific, falsifiable test for Reddy: a disclosed, segment-level revenue number attached to the growth pillar in the FY2026 or FY2027 results. Not narrative language, not "momentum," not partnership announcements. A line item. If it appears and is material, FORGE 2031's growth pillar is real. If FY2027 arrives with the pan-African marketplace still described only in prose, it belongs on the list with Private Placements.

That is the operational picture. The remaining question is what the market is paying for it.

↑ Back to Top

VIII. The Numbers That Matter: Growth, Margins, and a Global Valuation Gap

Here is a small puzzle. Take a company earning a 22% return on equity, growing revenue in the mid-teens, converting that to earnings growth in the high teens, paying out most of its profit, holding a near-unassailable position in its home market, and owning a stake in the country's only securities depository. Now ask what multiple of earnings the market assigns it.

The answer, through 2026, has been roughly eleven to twelve times.⁵⁷⁵⁸ JSE's shares traded at a trailing price-to-earnings ratio of around 11.1 in August 2026, with a dividend yield in the region of 6.2%.⁵⁷ On a ten-year view the stock has been remarkably consistent about this: its price-to-earnings ratio in mid-2026 sat about 6% below its own ten-year median of roughly 12.5.⁵⁸

For context — and these are approximate market observations that move daily rather than fixed facts — global exchange operators have generally traded in the low-twenties to low-thirties times trailing earnings through this period. London Stock Exchange Group, Nasdaq, Deutsche Börse, Hong Kong Exchanges and Clearing, and ASX have all commanded multiples roughly double to triple JSE's. Even Brazil's B3, an emerging-market exchange in a volatile currency with its own domestic political risk, has typically traded meaningfully above JSE. On enterprise-value-to-EBITDA, JSE has sat around seven times against a peer range roughly in the eleven-to-twenty band. JSE's dividend yield has exceeded every peer in the set, generally by a wide margin.

So the market is applying a discount of roughly half to two-thirds. The analytical question is what, precisely, that discount is compensating for.

The comparison that eliminates the easy answer

The lazy explanation is that this is an Africa discount, or an emerging-market discount, and no further thought is required.

That explanation runs into an inconvenient case. Nigeria's NGX Group — the operator of a smaller, materially less liquid exchange in a market with higher currency, political and macroeconomic risk than South Africa by almost any measure — has traded at roughly 37 times trailing earnings. It is possible to construct arguments about float, local investor base composition, and the peculiarities of Nigerian equity pricing that explain some of this. But the crude version of the emerging-market-discount thesis does not survive the comparison. If African exchange operators were simply de-rated as a class, NGX would not trade at three times JSE's multiple.

Which means the discount is specific. Something about JSE, not about its continent, is what the market is marking down.

There are three plausible candidates, and they map exactly onto the previous three sections.

The first is the growth ceiling implied by the listings pool. A high multiple is a claim about future growth. If investors believe South Africa's public market is structurally narrowing, they are implicitly capping how much value JSE can compound over a decade, regardless of how well it monetises today's activity. A toll collector on a road with declining traffic is worth less than a toll collector on a growing one, even at identical current margins.

The second is the legal overhang. An unresolved abuse-of-dominance referral seeking up to 10% of turnover, plus structural remedies to the rules protecting the group's second-largest segment, is not a quantifiable liability but it is an unpriceable one — and markets discount unpriceable risk heavily. Some portion of this discount is simply the market declining to underwrite a Tribunal outcome it cannot forecast.

The third is currency and country risk in the ordinary sense. A rand-reporting company with a rand-denominated dividend, whose most attractive segment is dollar-earning and therefore hurt by rand strength, presents a genuinely awkward proposition to a global investor. There is no natural currency in which JSE is a comfortable holding.

The two readings, and what separates them

The value case runs as follows. This is a high-return, cash-generative, capital-light infrastructure business with a documented shift toward recurring revenue, a compounding depository royalty that survives competitive share loss, a payout ratio target of 67–100% that returns most of what it earns, and a balance sheet strong enough to have funded both a special dividend and a share repurchase programme in 2026.⁸⁵⁹ Investors, on this view, have over-extrapolated the listings decline — despite fifteen years of evidence that JSE grows earnings through it — and are pricing the antitrust case as if the worst outcome were the likely one.

The value-trap case runs as follows. The market is not making an error; it is pricing four things correctly. Growth is capped by a genuinely narrowing pool of domestic issuers. The segment that drove the recent re-rating is cyclical, tied to a specific and reversible political tailwind, and has been guided by management itself to normalise. The best-quality segment is exposed to a currency that hurts it when the country does well. And the second-largest segment faces a legal challenge to the rules that protect it. On this view, eleven times earnings is not a discount; it is the correct price for a well-run business with a structurally limited runway.

Both readings are defensible on the evidence, and the honest position is that neither can be resolved from today's information. What can be identified precisely is what will resolve them: the Tribunal outcome, the market-cap-weighted listings flow once the 2026 one-offs clear, and the appearance or non-appearance of a disclosed revenue line under FORGE 2031's growth pillar. Those three data points, arriving over roughly eighteen to twenty-four months, will settle which story was right.

↑ Back to Top

IX. Bull vs. Bear: The Investment Debate

Every serious investment case has a spine — a specific mechanism by which the company wins from here, and a specific mechanism by which it breaks. Laying the facts side by side is not analysis. Stating which mechanism the evidence favours, and what would change that view, is.

Why JSE wins from here — and where each argument thins

The toll-collector argument. The strongest version of the bull case does not rest on trading at all. It rests on the observation that JSE has built a position at multiple points along a single pipeline — matching, clearing, settlement via Strate, back-office, data, indices — and that most of those tolls are collected regardless of who wins the execution fight. The Strate royalty in particular is indifferent to venue share, since rival exchanges settle through the same depository. This is the argument that best explains why JSE's earnings have compounded through a fifteen-year decline in listing counts.

Where it thins: the toll positions are not equally secure. Execution is contestable on price. Post-trade is contestable in law. Only the depository stake is genuinely structural, and it contributes roughly R192 million — meaningful, but not the majority of the case.⁷

The diversification argument. Non-trading income at roughly a third of the total, up from a minority, is real and measured, and it is the single clearest evidence that this management team executed a stated strategy rather than merely describing one.¹²⁸

Where it thins: as established, several lines counted as non-trading are activity-sensitive, and the two most genuinely acyclical lines carry their own specific exposures. Diversification has reduced cyclicality. It has not removed it, and describing it as structural insulation goes beyond what the disclosure supports.

The governance argument. The 2024 remuneration dissent and the board's subsequent response — engaging 78.4% of the register, tightening metrics, raising the ROE vesting hurdle from 19.5% to 23% — is documented evidence that this board responds to shareholder pressure with structural change rather than explanation.⁵⁴⁵⁶

Where it thins: one episode, well handled, is evidence of responsiveness under pressure. It is not evidence of a durable culture, and it sits alongside the amaBhungane transparency episode, where JSE's response to external pressure was to fight the Information Regulator's finding.²⁸

The listings reversal argument. Coca-Cola HBC and Canal+ are large, real, and directionally new.⁴⁴⁴⁵

Where it thins: both are M&A-driven secondary listings, not evidence of a domestic IPO recovery. They will add index weight and tradeable value — genuinely good for revenue — while leaving the structural issue Kamlana identified untouched.

The survivorship argument. Two challengers have now been tested against JSE's position. One is dead; the other, after nearly nine years, holds low single-digit share of executed value on the most reliable measures.

Where it thins: the surviving challenger is now suing, backed by a well-capitalised shareholder base, with a state competition authority carrying its argument. That is a materially different threat profile from a startup trying to out-execute an incumbent.

Why it breaks

The antitrust mechanism. If the Tribunal finds that JSE's share is sustained by BDA interoperability friction and asymmetric trade-type rules rather than by liquidity, then the moat as described above is partly a legal artefact — and legal artefacts can be removed by order. The remedy sought reaches beyond execution into clearing and settlement, exposing the R1,082 million post-trade segment, and the penalty sought could approach a year's profit.¹³ This is the single largest identifiable risk to the investment case, and it is unresolved.

The cyclicality mechanism. Equity trading revenue grew 28.5% in FY2025 on a surge in value traded that management attributes substantially to post-GNU sentiment.⁷¹² Political sentiment reverses. And management has already guided the H1 2026 momentum to normalise in H2.⁸ An investor buying JSE on trailing earnings is buying earnings the company has said will not repeat at that growth rate.

The pool mechanism. Even accepting Allan Gray's consolidation argument in full, a market with fewer new domestic issuers has fewer future large-caps. That is a slow variable, but it compounds against the company rather than for it, and no amount of trading-revenue strength changes its direction.

The conversion mechanism. Private Placements at R327,000 and falling; basis futures abandoned after four years because the technology could not scale and a competitor stood ready to take the business.⁷⁵³ A five-year growth strategy resting on a pan-African marketplace and data monetisation asks investors to extend credit against a documented record of poor conversion from strategic initiative to revenue.

The embedded headwinds. A stronger rand mechanically compresses the best segment; falling rates mechanically shrink JSE Investor Services. Neither is hypothetical — both showed up in the reported H1 2026 numbers.⁸

The activist lens, applied where it belongs

JSE Limited is not the subject of a public activist campaign, and manufacturing one for analytical effect would be dishonest. But the closest available proxies for skeptical large-shareholder pressure are informative.

The most useful proxy is the register itself. The 2024 remuneration vote showed that roughly a fifth of shareholders were prepared to object publicly and in writing, and that the board moved when they did — which establishes both that organised institutional pressure exists here and that it works. Alongside that, the company launched a share repurchase programme in 2026 under a mandate approved at the May AGM, having already paid a special dividend.⁵⁹ The precise composition of the top of the register is not restated here; what matters analytically is the demonstrated responsiveness, not the identity of any single holder.

A skeptical investor would press on three things. First, pay-for-performance rigor: given that the biggest driver of recent earnings was a market-wide trading surge management did not create, how much of incentive outcomes reflect skill versus beta? The ROE hurdle increase partly addresses this, but ROE also rises on a cyclical trading boom. Second, capital return versus reinvestment: a buyback plus special dividend alongside a five-year growth strategy is a slightly mixed message, and the honest reading is that it reflects a board with more cash than high-confidence investment opportunities — which is arguably the right call given the conversion record, but should be named as such rather than presented as ambition. Third, accountability architecture: Fourie was ultimately, publicly measured against the listings problem. FORGE 2031 needs targets specific and auditable enough to measure Reddy the same way, and as of the August 2026 presentation, the growth pillar did not have them.

The calibrated conclusion

Weighing it: the evidence leaves JSE's core franchise claim intact but narrowed. The company is genuinely a diversified market-infrastructure operator with multiple durable toll positions, not merely a trading venue — that much the fifteen-year earnings record through a shrinking listing count establishes. But the strongest form of the claim, that this is a structurally protected compounding franchise with permanent economics, is not supported. Two of its four legs are contestable: execution share on price, and post-trade economics in law. A third — trading growth — has been guided down by the operator itself.

The most useful way to hold the position is that JSE's durability has been demonstrated and its growth has not. The evidence for "this business survives and earns well" is strong and historical. The evidence for "this business compounds materially faster than the South African market" is thin and prospective.

↑ Back to Top

X. Durable Lessons: What Running (and Owning) an Exchange Actually Teaches

Strip away the South African specifics and this story contains four lessons that transfer to a much wider class of businesses.

An exchange's moat is a function of its customers' customers. JSE's actual product is not a listing or a matching engine; it is liquidity. And liquidity is a network property that can decay very slowly while the income statement looks excellent, because trading revenue is generated by activity in whatever companies remain listed, not by the health of the pipeline replenishing them. The two series can diverge for a decade or more before the lagging one catches up. Anyone analysing a marketplace business — an exchange, a payment network, a listings platform, a job board — should track supply-side replenishment separately from monetisation of the existing base, because monetisation is the flattering series and it is the one management reports first.

Diversification away from a cyclical core is a real defense, and it is not a cure. JSE genuinely reduced its dependence on equity turnover, and the numbers prove it. But no amount of data, index and clearing revenue addresses the question of why fewer South African companies want to be public. Incumbents facing a shrinking primary market frequently mistake successful adjacent monetisation for having solved the underlying problem. The two are unrelated; the first buys time to fix the second.

Rules can be legitimate safeguards and competitive weapons at the same time, and only an external adjudicator can separate them. Capital adequacy requirements for exchanges genuinely protect the market — ZAR X's collapse would have been far more damaging had it happened mid-trade with client assets exposed. Those same requirements kept a credible competitor out. System interoperability standards genuinely protect settlement integrity. Those same standards may be creating switching friction that entrenches an incumbent. There is no way to tell which is which from inside the incumbent's framing, which is exactly why an active competition inquiry into those rules is the right place to test a moat rather than the wrong one. Investors should treat a serious antitrust proceeding not merely as a risk to be discounted but as free, adversarial due diligence on the durability of the advantage they are underwriting.

Governance credibility is revealed by friction, not by its absence. JSE's governance record is not clean — a 22% pay revolt, an adverse Information Regulator finding, a five-and-a-half-hour outage on a thirty-five-year-old system, two of the country's largest frauds on its board. What is informative is the response pattern: it raised its own ROE vesting hurdle when shareholders objected, funded a modernisation programme for the failed system, and fought the transparency finding. That is a specific and mixed profile — responsive on capital and pay, defensive on disclosure. It is far more analytically useful than a company with no controversies, because a company with no controversies has provided no evidence of how it behaves under pressure.

↑ Back to Top

XI. Epilogue: What to Watch From Here

Three things will decide which reading of JSE Limited was correct, and none of them require an investor to forecast a market.

The Tribunal. The outcome of the Competition Commission's abuse-of-dominance referral, and specifically whether any remedy compels changes to BDA interoperability or the matched-principal rules. A dismissal validates the moat in the strongest available way. An adverse structural remedy removes friction protecting the group's most profitable order flow and reaches into post-trade economics.¹³

Market-cap-weighted net listings flow. Not the headline count — the count is a poor proxy, as the gold-mining consolidation demonstrates. The measure that matters is the market capitalisation arriving versus departing, and specifically whether it stays positive after the Coca-Cola HBC and Canal+ secondary listings have cleared the 2026 comparatives.⁴⁴⁴⁵ That is the test of whether the Simplification Project, the segmentation reform and Operation Phumelela produced a structural change or a good year.

A revenue line under FORGE 2031's growth pillar. Whether the pan-African digital marketplace and data-and-services commercialisation produce a disclosed, material, segment-level number in the FY2026 or FY2027 results — or whether they join Private Placements and the suspended basis futures contracts on the list of strategically-framed initiatives that never scaled.⁸⁷⁵³

Return to where this started. In March 2026 an exchange founded beside a gold rush reported the best profit in its 139 years, paid a special dividend, and watched its chief executive leave for the sea. It did so while operating a market that has lost more than half its companies in three decades, in a country whose own financial regulator has said out loud that this is a problem.

That is not a contradiction waiting to resolve itself. It is the entire investment debate, stated as compactly as it can be stated. The next eighteen to twenty-four months of Tribunal rulings and listings flow will show which half of it was the signal.

JSE Limited

Table of Contents

JSE Limited: Can Africa's Oldest Exchange Survive the Companies It Made Rich?

I. Cold Open & Roadmap

II. From Gold Rush to Public Company: A Compressed History (1887–2006)

The exchange that became a stock

III. Building the Modern Machine: SAFEX, BESA, Strate, and the Logic of Vertical Integration (2001–2022)

2001: SAFEX, and the discovery that derivatives are better than equities

2009: BESA, and a documentation gap worth admitting

Strate: the quiet royalty nobody talks about

JSE Clear (2022): making a licence into a line item

The technology decision: renting the engine

IV. How JSE Actually Makes Money Today (Segment Deep Dive)

The group picture

Capital Markets: the engine, and the cycle

Post-Trade Services: the second pillar, and the legal exposure

Information Services: structural growth, currency friction

JSE Investor Services: the one going backwards

The diversification claim, and how much of it survives scrutiny

JSE Private Placements: optionality, sized honestly

V. The Moat on Trial: Industry Structure, A2X, and the Competition Tribunal

What an exchange moat actually is

A2X: the challenger that did not go away

Market share: state the evidence, including where it conflicts

The case itself, and why it is the right test

ZAR X: the cautionary tale that cuts both ways

Governance as competitive asset — and where it has frayed

VI. The Listings Exodus: South Africa's Shrinking Public Market

Myth versus reality: how bad is the shrinkage, actually?

The flow that actually matters

The structural case studies: why companies leave, and one that didn't

The reversal that is actually new

The policy response, tracked over time

VII. Leadership in Transition: Fourie's Record and Reddy's First Test

Leila Fourie: what the record actually shows

Remuneration: pay, and what happened when shareholders objected

Valdene Reddy: continuity, with everything still to prove

FORGE 2031: the plan, and what would prove it

VIII. The Numbers That Matter: Growth, Margins, and a Global Valuation Gap

The comparison that eliminates the easy answer

The two readings, and what separates them

IX. Bull vs. Bear: The Investment Debate

Why JSE wins from here — and where each argument thins

Why it breaks

The activist lens, applied where it belongs

The calibrated conclusion

X. Durable Lessons: What Running (and Owning) an Exchange Actually Teaches

XI. Epilogue: What to Watch From Here

References