In debates about the shrinking number of companies on the Johannesburg Stock Exchange, explanations often come easily. Some blame regulation, others weak economic growth, and still others political uncertainty.
But according to Professor Haroon Bhorat of the University of Cape Town, the reality is more complicated.
During a media briefing this week on a new research paper examining JSE delistings, Bhorat described the study as the product of a long investigative process. Researchers had to wrestle with fragmented global exchange data and construct a dataset capable of properly measuring delisting dynamics.
That effort has produced one of the most comprehensive analyses yet of why companies leave South Africa’s main exchange – and what the trend reveals about the health of the country’s capital markets.
The paper, Vanishing Acts: An Econometric Exploration of Firm Delistings in South Africa, was authored by Bhorat together with Leigh Neethling and Ayesha Sayed of UCT’s Development Policy Research Unit.
According to the researchers, no comparable study has previously been conducted for an emerging market. Although a handful of similar analyses exist for developed economies, few combine global benchmarking, econometric modelling, machine-learning techniques, and qualitative interviews in a single framework.
Beyond its findings, the research establishes a baseline dataset covering more than two decades of firm-level and macro-economic information – a resource the authors say could support further academic work on capital markets in emerging economies.
Most importantly, it challenges several assumptions that have dominated public discussion about JSE delistings and highlights the need to think more carefully about how listing trends are measured.
Between anecdotes and evidence
The decline in the number of JSE-listed companies has long been explained through competing narratives.
One view emphasises domestic policy and regulatory pressures. Another attributes the trend to global shifts in capital markets, including the rise of private equity and mergers as alternatives to public listings.
Until recently, however, much of this debate relied more on anecdote than systematic evidence. Sunette Mulder, chief of staff at the Association for Savings and Investment South Africa (ASISA), said this was precisely why the organisation commissioned the research.
“The lack of research into the drivers of JSE delistings often led various parties to develop their own theories,” she said, noting that these explanations were often used to support competing agendas in public debate.
The study therefore set out to provide an independent empirical analysis of the issue.
The result is a detailed examination of listing dynamics on the JSE and a dataset that researchers hope others will build on in future.
A unique dataset – and a new way of thinking about delistings
Constructing the analysis required assembling a dataset that did not previously exist.
The researchers combined firm-level financial variables – including market capitalisation, profitability and valuation ratios – with macro-economic indicators such as GDP growth, exchange rates, and policy uncertainty indices.
The resulting dataset covers firm-level observations from 2002 to 2024, while descriptive trends extend back to 1989.
Beyond the findings themselves, Bhorat argues the study highlights something else: the importance of thinking carefully about how delistings are measured.
Public discussion often focuses simply on the number of companies leaving the exchange. But that approach ignores the broader dynamics of listings and delistings occurring simultaneously.
“If you see a company delisting,” Bhorat said, the key question should be: what is happening on the other side of the balance sheet?
How many companies listed during the same period?
And what happens if the analysis shifts from simple counts to market capitalisation, which reflects the economic weight of firms entering and leaving the exchange?
Looking at those broader measures changes the picture considerably.
Delistings in context: what the global comparison shows
The study’s international comparison reveals that the JSE has experienced persistently negative net listing rates relative to many global peers.
However, the difference is strongly influenced by historical trends.
Much of the divergence occurred between the mid-1990s and early 2000s – a period shaped by South Africa’s democratic transition following the 1994 election, which ended apartheid and reopened the economy to global markets. Financial liberalisation and increased international participation were followed by a surge in inward listings on the JSE and, later, a cycle of corporate restructuring that contributed to elevated delisting activity in subsequent years.
Since then, the gap between the JSE and global exchanges has narrowed.
Why companies delist
One of the clearest findings concerns the reasons companies leave the exchange.
Using JSE data on delisting causes from 2007 onward, the researchers found that schemes of arrangement account for close to 50% of delistings. These transactions typically reflect mergers, acquisitions, or corporate restructurings rather than financial distress.
Other reasons are suspensions, compliance failures, and administrative removals.
The finding suggests that much of the contraction in the number of listed firms reflects structural corporate activity rather than collapse.
The rise of the large-cap exchange
Another structural trend revealed by the study is the growing dominance of large companies on the JSE.
Using the Herfindahl-Hirschman Index – a standard measure of market concentration – the researchers found that concentration on the exchange has risen significantly over the past two decades.
As smaller companies leave the exchange, large-cap firms account for an increasing share of market value. This shift has implications for liquidity, analyst coverage, and investor behaviour, particularly for smaller firms that may struggle to attract capital in an increasingly concentrated market.
A closer look at mining
Mining has often featured prominently in discussions about JSE delistings, with regulatory changes – particularly the Mining Charter – frequently cited as a major driver of companies leaving the exchange.
The researchers therefore examined the sector more closely.
At first glance, the data appears to support the narrative. Delistings among basic materials companies increased between 2019 and 2022, overlapping with the implementation of stricter transformation requirements under the 2018 Mining Charter.
However, a deeper analysis disclosed a more nuanced picture.
Many transactions initially interpreted as companies leaving the JSE were corporate restructurings rather than genuine exits. Several involved dual-listing reorganisations or structural changes while firms remained listed.
Once these cases were removed, the scale of the impact changed significantly.
Early estimates suggested that mining-related delistings linked to the Charter represented market-capitalisation losses of more than R1 trillion. After adjusting for restructuring events that did not represent true exits, the actual loss attributable to the charter fell to about R46 billion – roughly 0.24% of the JSE’s market capitalisation.
This does not mean regulation had no impact. Instead, the findings suggest the effects may have been more concentrated among smaller and junior mining companies, which face greater compliance and financing constraints than larger firms.
What the models show
To move beyond descriptive trends, the researchers estimated the probability that firms would delist using several econometric and machine-learning models.
Across these approaches, the results were consistent.
Companies were significantly less likely to delist if they were:
- larger in market capitalisation,
- older firms,
- more profitable, and
- carrying higher leverage.
Conversely, smaller and less profitable firms faced higher exit risks.
Macro-economic conditions also played a role.
Stronger South African GDP growth reduced the probability of delisting, while stronger US economic growth increased it, suggesting global capital markets can pull firms towards opportunities abroad.
Taken together, the results highlight the importance of firm fundamentals and broader economic conditions in shaping listing decisions.
What the data cannot show
Even with its extensive dataset, the study highlights important limitations.
One question remains fundamentally unanswerable: how many companies might have listed but never did.
The dataset can analyse firms that entered and exited the exchange, but not companies that chose to remain private or pursue alternative funding routes.
Listing decisions are highly sensitive to timing and market conditions. Firms often go public when investor sentiment is strong, valuations are favourable, and capital markets are liquid. When conditions deteriorate – whether because of weak growth, policy uncertainty, or subdued investor appetite – companies may simply postpone or abandon listing plans.
Other deterrents can also play a role, including compliance costs, disclosure requirements, and concerns about liquidity or analyst coverage.
In other words, some of the most important dynamics shaping the exchange occur before a company ever reaches the listing stage.
Understanding those “missing listings” may therefore be just as important as understanding delistings themselves.
The big dial: economic growth
Ultimately, the research points to a conclusion that may sound simple but carries important implications.
The most powerful determinant of listing activity is not regulation, sentiment, or policy debates – it is the underlying strength of the economy.
As Bhorat put it during the briefing, the “big dial” determining whether companies list and remain listed is economic growth.
Exchange-level reforms may help at the margins. But sustained improvements in capital market depth depend on stronger investment, business formation, and economic expansion.
A baseline for the future
Mulder says the study provides something South Africa’s capital markets have never had before: a baseline measurement of listing dynamics.
The research ran through to 2024, just before the JSE introduced its Listings Simplification Project in 2025.
In a few years, analysts will be able to revisit the dataset and assess whether those reforms had a measurable impact on listing behaviour.
Having a baseline now makes that future analysis possible.
The challenge of changing the narrative
One question during the briefing captured the broader implications of the findings.
If nearly half of delistings reflect mergers and restructuring rather than collapse, could the research help to restore confidence in the exchange?
Bhorat’s response was cautious.
Whether the findings change public sentiment depends on how widely the evidence is understood. Human behaviour, he suggested, often favours stories over statistics.
He recalled a remark from a former Mexican deputy finance minister who once observed during cabinet meetings that “anecdote beats regression analysis every time”.
Economists can produce rigorous models and detailed datasets. But a single story – for example, a mining company deciding to shut operations after receiving a letter from the Ministry of Mineral Resources and Energy – can still dominate the public conversation.
And so, the debate continues – between anecdotes and evidence, narratives, and data – over what the future of South Africa’s capital markets will look like.
Read the full working paper.
