The Naspers/Prosus Story: The only bill sent was for winning
In the middle of two thousand and one, two investors sold out of a loss making Chinese messaging company.
One was IDG, an American venture firm. The other was PCCW, a Hong Kong telecoms group. Each had paid about two point two million dollars for roughly a fifth of the business. Both were selling. The dot com crash had just destroyed the market for companies like this one.
The buyer was a South African newspaper company that had already burned a hundred million dollars failing in China, and was preparing to go home. It paid thirty two million dollars.
That stake would peak at an estimated one hundred and eighty to two hundred billion dollars.
THE PARTY’S PRESS
On the twelfth of May, nineteen fifteen, a company called De Nationale Pers Beperkt was registered in Cape Town. National Press Limited. In time the name would be shortened to Naspers.
It had been agreed some weeks earlier, at the home of a district bank manager named Hendrik Bergh in Stellenbosch, where sixteen Cape Afrikaners resolved that a defeated language needed a printed voice. The driving force was an attorney, William Angus Hofmeyr. The money came from two wealthy Stellenbosch landowners, Jannie Marais and his brother Christiaan.
On the twenty sixth of July that year, the first edition of De Burger appeared. Its editor was D.F. Malan. Thirty three years later, Malan would lead the National Party to power and begin legislating apartheid.
That is not a coincidence discovered later. It is the origin. Naspers was not a business that afterwards picked up a political position. It was a political project that happened to be organised as a company.
The early problems were real, and they were not the problems of a normal business. There was almost no capital — the company lived on subscriptions from a poor readership and the patronage of a handful of Cape farmers. And the market itself barely existed. Afrikaans was not a fully standardised written language in nineteen fifteen, and would not be recognised as an official language until nineteen twenty five. This was a publisher selling to a readership it partly had to invent. Die Huisgenoot followed in nineteen sixteen, and became the most widely read Afrikaans magazine in the country.
Then, in nineteen forty eight, the man who had edited the first newspaper became prime minister. For the next forty six years the National Party governed South Africa, and this company’s titles were the movement’s house press. Its readership and the governing party were, for practical purposes, the same people. The politics succeeded, and because the politics succeeded, the business succeeded.
In nineteen ninety four, apartheid ended, and the purpose the company had been founded to serve ended with it.
In nineteen ninety seven, the Truth and Reconciliation Commission asked South Africa’s media companies to account for their role between nineteen sixty and nineteen ninety four. Naspers declined to submit. Its rivals, Afrikaans language and English language alike, engaged.
A hundred and twenty seven of its own journalists submitted a separate representation. They apologised. They wrote that the company’s newspapers had formed an integral part of the power structure which implemented and maintained apartheid. Management was not among the signatures.
The corporate apology came eighteen years later, in July twenty fifteen, at the company’s own centenary celebration, delivered by the chief executive of Media24, Esmaré Weideman. No restitution was attached to it. South African commentary called it a half apology.
And nothing followed. No penalty, no restitution, no cost of any kind. The company that declined the Truth and Reconciliation Commission in nineteen ninety seven went on to become the most valuable company on the continent.
Many South African corporations were complicit in apartheid. What is unusual about this one is not the complicity. It is that it paid nothing.
Nineteen ninety seven was also the year Naspers handed itself to a new chief executive. His name was Koos Bekker, and he had spent the previous decade building something nobody at the newspaper had thought was a media business at all.
NO NEWS
Koos Bekker grew up on a farm near Heidelberg, in the Transvaal. He took degrees in law and in literature, and then, in the early nineteen eighties, he sold his house and moved to New York to study for an MBA at Columbia Business School.
What he watched from there was the rise of HBO. Americans were paying a monthly subscription for television. He wrote his thesis on whether the same idea could work in South Africa, and worked it up with two friends, Cobus Stofberg and Jac van der Merwe.
In nineteen eighty four, the three of them pitched it to the chief executive of Naspers, Ton Vosloo. The argument they made was not visionary. It was defensive. Naspers was a print company, and it was losing advertising revenue to state television. A subscription service would take its money from viewers instead of advertisers, which meant it would never have to win that fight at all.
Vosloo said yes.
In nineteen eighty five a consortium of South African print publishers — Naspers among them, alongside Argus, Times Media and Perskor — was awarded the country’s first pay television licence, for a service called M-Net. It went on air in October, nineteen eighty six. The licence carried a condition. M-Net could broadcast no news. News belonged to the state.
So the company founded to publish newspapers was granted a television licence by the government its newspapers had helped put in power, on the condition that it not report the news. Nobody had to fight for that licence. It was granted, on the state’s terms, to the state’s own press.
Naspers took a twenty six per cent stake and put up the launch funding. In nineteen ninety three the subscriber management arm was separated out as MultiChoice. In nineteen ninety five the satellite service DStv launched.
And it worked. M-Net worked, MultiChoice worked, DStv worked, and the man who had written the thesis made himself and Stofberg rich doing it.
In nineteen ninety seven, Bekker became chief executive of Naspers itself.
Two years later he made an arrangement almost no chief executive of a listed company has ever made. From nineteen ninety nine, for roughly fifteen years, he took no salary. No bonus. No benefits. Nothing. He was paid exclusively in share options that vested over long periods.
Across his time as chief executive, the company’s market value went from roughly six hundred million dollars to about forty five billion. The arrangement made him a billionaire, and it is the single most cited fact about him.
It had a quieter effect too. A chief executive who is paid only in options that vest years from now has no reason to care what happens in the next quarter.
THE BUYER OF LAST RESORT
By the late nineteen nineties, Naspers had built an offshore vehicle for internet and emerging market investments, called MIH, Myriad International Holdings. Antonie Roux ran the internet operations. The idea was to find the next M-Net somewhere outside South Africa, and China was where they looked.
It went badly. Roughly a hundred million dollars went into Chinese ventures that failed. Not one venture, and not cheaply. By two thousand and one the China strategy had produced nothing, and the team was preparing to wind it down.
In July, two thousand, an MIH business development executive named David Wallerstein was walking into internet cafés in Chinese cities. In the standard telling, what he noticed was that on nearly every screen in every café, the same small penguin icon was running. It was an instant messaging service called QQ. He tracked down the company that made it, in Shenzhen, and made contact.
The company was called Tencent. It was losing money, it had no meaningful way of making any, and the market for companies like it had just been wiped out.
In May and June of two thousand and one, MIH bought forty six and a half per cent of Tencent for about thirty two million dollars. Charles Searle, the investment executive who did the deal, joined the Tencent board that June. He is still on it, twenty five years later.
The sellers were IDG and PCCW. Each had bought roughly a fifth of Tencent for about two point two million dollars. Each was getting out.
There was no auction here. There was no competition for this asset. There was a Shenzhen company burning cash, two shareholders looking for an exit, and one buyer left in the room — a South African newspaper company that had already lost a hundred million dollars in that country and had almost nothing to show for it.
There is a competing account, in which IDG shopped the deal to MIH rather than Wallerstein finding it in a café. Both are probably partly true, and it does not much matter which, because neither version disturbs the two facts that carry this story: who was selling, and why.
So the famous number is real. It just does not mean what it is usually taken to mean. Naspers was not the smart money in that room. It was the residual money. Thirty two million dollars was not the opening move of a visionary campaign — it was close to the last cheque of a losing one.
SEVENTEEN YEARS
In two thousand and four, Tencent listed on the Hong Kong Stock Exchange. What Tencent went on to become is a different story. What matters here is that the value of what Naspers owned kept multiplying, year after year, for seventeen years.
And Naspers did not sell.
Not in two thousand and four, when the listing finally gave the stake a public price. Not in two thousand and eight. Not in twenty ten, or twenty thirteen, or twenty fifteen. The single most consequential thing this company did across those seventeen years was nothing at all.
That is harder than it sounds. Holding an appreciating asset in another hemisphere, in a business you do not operate, under a currency and a government you do not control, is exactly the concentration risk that every framework in finance exists to prevent.
So who could have made them sell?
Nobody could. Naspers has two classes of share. The N ordinary shares carry one vote each, they are listed in Johannesburg, and they are what the public owns. The A ordinary shares carry a thousand votes each. They are unlisted, they carry almost no economic participation, and they are held through two companies — Keeromstraat thirty Beleggings and Naspers Beleggings. Between them, they control more than half the votes in the company.
Koos Bekker holds some. Cobus Stofberg holds some. The insurer Sanlam holds some. The beneficial holders of the majority are not publicly known.
The consequence is plain. Every resolution at every annual general meeting passes. The rival publisher Caxton has campaigned against the structure for years. Fund managers have said publicly that shareholders here have no real say. South African savers own this company through their pension funds and cannot vote on any part of it.
That is indefensible, and it is also the entire reason there is a story to tell. No shareholder base could force a sale, so no shareholder base did. Put beside it a chief executive with no salary to protect and no quarter to answer for, and you have the only conditions under which anybody holds an asset like this for seventeen years. The governance and the outcome are not two facts sitting awkwardly next to each other. They are one fact.
Then, in March twenty eighteen, Naspers sold two per cent of Tencent for nine point eight billion dollars. The first time in seventeen years it had sold any of it. Nobody had the votes to make them do that.
The stake bought for thirty two million dollars peaked at an estimated one hundred and eighty to two hundred billion. Roughly five thousand times what was paid for it. The most profitable venture investment ever recorded.
THE ONLY BILL
By twenty eighteen, Naspers was roughly a quarter of the entire Johannesburg Stock Exchange.
Not a quarter of the media sector. A quarter of the market.
South African institutional mandates cap how much of a single company any one fund may hold. So a domestic pension fund watching this stock grow to a quarter of the index it is measured against had no decision to make. It sold. It sold the best performing asset it owned, and then it sold more, because the better the stock did the further out of compliance the fund went. Every domestic fund in the country was structurally obliged to sell Naspers, whatever any of them thought of it.
Nobody could force Naspers to sell. Arithmetic could.
That is also where the discount came from. Both entities came to trade far below the value of the assets they held — a discount to net asset value, in the language of the market, and a wide and stubborn one.
And that discount was not a verdict. No fund manager in Johannesburg was marking this company down for nineteen forty eight, or for refusing the Truth and Reconciliation Commission, or for the A shares. They were rebalancing. The market has no memory and no opinion. It had simply become arithmetically impossible for South Africa to hold as much Naspers as Naspers was worth.
A company that had never once been made to pay for anything it did was finally being charged — for winning.
The responses came in order. February twenty nineteen: MultiChoice was unbundled, handed to Naspers shareholders and listed separately in Johannesburg. The pay television business Bekker’s thesis had created left the group.
On the eleventh of September, twenty nineteen, Prosus listed on Euronext Amsterdam: the international assets and the Tencent stake, moved into a deeper capital pool. Naspers kept the majority. The discount narrowed, and then it persisted.
In April twenty twenty one, Prosus sold another two per cent of Tencent, for fourteen point six billion dollars.
Then, in August twenty twenty one, came the cross-holding. Naspers and Prosus swapped shares until Prosus owned about forty nine per cent of its own parent, while Naspers owned somewhere between fifty seven and sixty per cent of Prosus. Each company now owned a large piece of the other. The purpose was to shrink Naspers’s weight on the Johannesburg exchange by shifting value into the Amsterdam entity.
Investors hated it. It was circular, it was opaque, and it did not work. After the cross-holding, Naspers’s discount to net asset value ran at around fifty six per cent. The company was valued at less than half of what it held. The engineering had made worse the exact thing it was built to fix.
In June twenty twenty three, Naspers and Prosus announced their intention to remove the cross-holding. Shareholders approved it in August and it was implemented in September. And on the eighteenth of September, twenty twenty three, three weeks after the structure he had built was taken apart in public, the chief executive of both companies, Bob van Dijk, resigned.
Underneath all of it ran the buyback. From June twenty twenty two, the group sold Tencent shares and bought back its own, on the logic that buying your own assets at half price is the best trade available to you. Cumulatively, forty six billion dollars. The Tencent stake came down from thirty three point two per cent toward twenty three.
A spin off, a listing in another country, a structure that had to be dismantled in public, a chief executive gone, and forty six billion dollars spent. All of it aimed at a bill that nobody had sent, and that nobody could be asked to withdraw.
THE BRAZILIAN
On the first of July, twenty twenty four, Fabricio Bloisi became group chief executive of both Prosus and Naspers. He is Brazilian. He founded Movile, and then iFood, the food delivery company that is now the group’s most valuable operating asset. The head of the group is the founder of the group’s crown jewel.
He rewrote what the company says it is. Not a holding company that owns things any more, but an operator of what he calls artificial intelligence driven lifestyle ecosystems, across Latin America, India and Europe.
And in the financial year to the thirty first of March, twenty twenty six, it worked. Ecosystem revenue of nine point seven one billion dollars, up more than half. Adjusted core earnings up eighty four per cent. Record free cash flow. The dividend raised by forty per cent. And for the first time ever, every regional ecosystem in the group was profitable.
The best operating year in a hundred and eleven years.
On the day we are telling this story, Naspers and Prosus are both sitting at fifty two week lows, down roughly thirty per cent over twelve months. The reason is Tencent, down about twenty six per cent this year as investors take fright at what artificial intelligence is costing it. Eight years, forty six billion dollars, an Amsterdam listing and a Brazilian chief executive later, these shares still move on a chart drawn in Shenzhen.
And in Brazil, the attack has already begun. Meituan, the Chinese delivery giant, entered the country under the brand Keeta with five billion dollars committed over five years. DiDi relaunched its Brazilian service, ninety nine Food, undercutting iFood on what restaurants pay. Brazil’s competition authority, CADE, prohibited the exclusivity contracts iFood held with the big restaurant groups, which had been its moat. And on the twelfth of May, twenty twenty six, Bloisi told the market that iFood’s earnings in twenty twenty seven would be lower, because of what defending Brazil is going to cost. Both stocks fell that day.
The company that got rich renting a seat on China’s internet is now being attacked by China’s internet, in the one business it owns outright. How that ends, nobody knows, and that includes us.
Back home it is quieter. Phuti Mahanyele-Dabengwa runs what is left of Naspers in South Africa. Takealot, the online retailer, turned its first full year profit and passed a billion dollars in local revenue. Media24, the direct descendant of the nineteen fifteen company, has been shutting its presses. Rapport printed its last edition on the twenty second of December, twenty twenty four.
And in Cape Town, Die Burger, the newspaper this company was founded in nineteen fifteen to print, is still printing. Its circulation is twenty nine thousand one hundred and seventy three.
Prosus is Dutch, primary listed in Amsterdam, run by a Brazilian, and growing in Brazil, India and Europe. Naspers is the Johannesburg parent of an online retailer and a publishing arm being wound down to one printed newspaper. So: is this still an African company? Nobody in the group has answered that directly, and we are not going to answer it for them. This is Asili Africa. Every empire has an origin. See you on the next one.
Key Takeaways
- THE PARTY’S PRESS. On the twelfth of May, nineteen fifteen, a company called De Nationale Pers Beperkt was registered in Cape Town.
- THE BUYER OF LAST RESORT. By the late nineteen nineties, Naspers had built an offshore vehicle for internet and emerging market investments, called MIH, Myriad International Holdings.
- SEVENTEEN YEARS. In two thousand and four, Tencent listed on the Hong Kong Stock Exchange.
- THE BRAZILIAN. On the first of July, twenty twenty four, Fabricio Bloisi became group chief executive of both Prosus and Naspers.
Also available on YouTube — search “Asili Africa” or subscribe to our channel.