R27 billion can fund one of the following: -2000 to 4000 electric busses, -20 % of the Gautrain Expansion -160 new blue train sets (800 train cars) -25 000+ bus driver salaries for 4 years .....or...... 5 months of mildly cheaper petrol (yes, we chose cheaper petrol)

Loose Canon 💥 A Nobel for criticising the economics of saying nothing at great length timcohen.co.za/loose-canon-a…

President Cyril Ramaphosa has criticised South Africans seeking refuge in the US, saying he finds it “unbelievable” that some continue to believe the falsehoods about the country, including claims of a white genocide. Video: Thando Radebe

#MustRead The Cover-Up: How a Trillion-Dollar Industry Launders Illegal Gold. "The difference between “clean” and “dirty” gold is a fiction. There is just gold. A Times investigation found that the London Bullion Market Association, which banks and governments rely on to police the market, repeatedly allowed the sale of gold from criminals and cartel members. In Madagascar, where thousands of children work in illegal mines, the association helped change and backdate key paperwork that certified gold as free from abuse." nytimes.com/2026/10/19/world… via @NYTimes

"KOMBINOWAĆ": The Polish art of outsmarting reality. There is a Polish verb that has no satisfactory equivalent in English or French. It is kombinować, and understanding it is essential to understanding the Polish national character. Kombinować means finding a solution to a problem through ingenuity, improvisation, personal connections, dubious arrangements, and occasionally methods that would make a German lawyer suffer a nervous breakdown. It is neither necessarily legal nor necessarily illegal. Neither entirely honest nor entirely dishonest. It occupies that fascinating territory between human intelligence and criminal entrepreneurship. And Poles have elevated it to something approaching an art form. Imagine that three Europeans discover their washing machines have broken down. The Englishman telephones a repair company. The technician arrives on Tuesday, identifies the defective component, orders a replacement, and schedules another appointment. The Frenchman telephones customer service, argues for forty minutes about consumer rights, threatens legal action, and eventually writes an indignant letter to the manufacturer. He might even establish an association of dissatisfied washing-machine owners. The Pole looks at his washing machine and announces: Spokojnie. Coś wykombinujemy. (Don't worry. We'll "figure something out", but it's a bit more than that...) What follows is difficult to describe within the conventional boundaries of economic science. He telephones his cousin, who knows a mechanic, whose brother-in-law once worked at a factory producing agricultural equipment. The mechanic has no washing-machine components, but he possesses a motor from an old Soviet tractor. The motor is too large. No problem. A neighbour knows somebody who can modify it. Unfortunately, that somebody owes money to another neighbour, who happens to need help transporting a refrigerator. Three telephone calls, two bottles of vodka, one borrowed trailer, and an entirely unrelated dispute about a missing bicycle later, the washing machine works. It now sounds like a MiG-21 preparing for takeoff, but it works. And the Pole is delighted. He has not merely repaired an appliance. He has defeated reality. This is kombinowanie. The Polish “Coś wykombinujemy” contains something more adventurous than "we'll figure something out". The process itself is an expedition into the unknown. French offers several partial equivalents. Se débrouiller conveys resourcefulness. Bricoler describes improvisation. Magouiller suggests suspicious arrangements. Combiner can refer to devising schemes. But Polish conveniently compresses all these possibilities into one verb. A man who kombinuje might be an ingenious inventor, an exceptionally resourceful neighbour, a slightly dishonest businessman, or a future defendant in a corruption trial. Sometimes all four. And this peculiar mentality has historical foundations. For much of the twentieth century, particularly under communism, Poles inhabited an economy in which official procedures frequently proved incapable of satisfying elementary material needs. You wanted a refrigerator? The shop had none. You wanted building materials? The warehouse was empty. You wanted to repair your car? The necessary components were unavailable. But somewhere, somebody knew somebody. The communist state had developed an elaborate system for planning production, allocating resources, and distributing goods. The Polish population responded by developing an equally elaborate system for circumventing it. An entire parallel economy emerged, built upon favours, exchanges, informal obligations, ingenious substitutions, and personal relationships. The bureaucrat possessed the regulations. The kombinator possessed his cousin's telephone number. And quite frequently, the cousin proved more useful than the entire Ministry of Industry. Naturally, there was a price to pay. A society organised around kombinowanie may produce extraordinarily resourceful individuals while tolerating spectacularly dysfunctional institutions. Why demand that the system function properly when you can always find a way around it? Why repair the road when you know a shortcut through somebody's field? Why simplify the administration when everyone knows that the appropriate stamp can be obtained through a friend? The kombinator triumphs over dysfunction, but his triumph can also help perpetuate it. Yet it would be a mistake to regard kombinowanie merely as the cultural residue of communism. The word and the attitudes it describes have a broader history. They express a particular admiration for intelligence that refuses to remain confined within established procedures. In this sense, kombinowanie has an almost artistic dimension. The engineer asks what an object was designed to do. The kombinator asks what else it could possibly do. The administrator sees a regulation. The kombinator sees an interesting literary text open to interpretation. The economist sees an inefficient allocation of resources. The kombinator sees an opportunity to exchange a broken motorcycle for six chickens, a barrel of diesel, and an introduction to a man who works at the town hall. There is something profoundly creative in this attitude, even when the results are absurd. Of course, the English also improvise, the French also circumvent rules, and the Germans are perfectly capable of inventive dishonesty. These are not exclusive national behaviours. What is distinctive is the extraordinary semantic and cultural career of the Polish word: its ability to turn practical improvisation, cunning, opportunism, and social resourcefulness into variations of the same recognisable activity. And perhaps the greatest difference lies in the attitude towards the unexpected. For the English pragmatist, a good solution is one that works reliably. For the French intellectual, a good solution should ideally also be theoretically defensible. For the Polish kombinator, a good solution is one that nobody would have imagined in the first place. Preferably one that also leaves him with a spare tractor motor, three bottles of vodka, and a neighbour who owes him a favour.

Mustek said its Business AI arm, in which it invested R8 million, has been forced to pivot as the South African market is not as ready for the technology as initially believed. mybroadband.co.za/news/busin…

Griquas winning back-to-back Currie Cups isn't spoken about enough Not even the Brumbies, Highlanders, Hurricanes or any NPC team have achieved this feat when you consider the fact that Super Rugby is on par with the Currie Cup

I have made a good-faith effort to learn from MMT and I can report that I learned nothing. If I was supposed to learn that a country that issues its own currency can inflate away its debt, I already knew that. (And I worry about that.)

Now your agent can trade perps for you on Coinbase Send “Connect to coinbase.com/skill.md" to your favorite LLM to get started

I can't imagine that this will end well. dailyinvestor.com/investing/…

I recorded a podcast with Tanner Greer (@Scholars_Stage) about this. m.youtube.com/watch?v=V1cwOW…

SUI Appointment of Shirley Machaba as an independent non-executive director and changes to certain committees bit.ly/4ioTzVD

The Durban Test had harsh lessons for cricket in the country in more ways than one cricinfo.com/story/south-afr…

I like this article that speaks about replacement ratios because most articles that refer to the tax benefits of pension contributions and retirement annuities never complete their articles. About 95% of them never run the rule for drawdowns And surprise surprise most of these articles pop-up in February in the month that the tax year and closes, almost an emotional appeal to a taxpayers desperation to reduce their tax. I would say 60% to 70% of them If you earn over R2.5m per annum and you use an income replacement rate of 75%, you will be earning over R1.875m in retirement. You were in the 45% tax bracket, you are still in the 45% tax bracket. What is the “tax benefit” of contributing to an RA? 🤔💭

Really great post. Bottom line: "Models are superintelligent where we have enormous data, and progressively less so as we have less data." Nice summary: "All the rules-of-thumb about learning are at the root, the same -- Malcom Gladwell's 10,000 hours, Moravec's Paradox, Scaling Laws, the Bitter Lesson, Age of experience. If your learning system (model, brain, whatever) is unconstrained by compute, intelligence is proportional to accumulated experience."

Here are four exchange control stamps in a South African passport from the mid-1980s, allowing the passport holder to take a particular amount of currency out of the country. These exchange controls were imposed by the South African government to restrict the outflow of capital from the country.

Interesting read. Same Warren Wheatly From (BAC) African Bitcoin Corporation (formerly Altvest) Lebashe was also a major shareholder in EOH. Company that sunk with corporate corruption & has since been renamed iOCO Sums up BEE Black Elitist Enrichment billionaires.africa/2026/09/…

I spent twenty years in anti-money-laundering compliance at one of the big American banks. I will not say which, because it does not matter which; we all run the same machine, and this week it was Bank of America’s turn to open the hood, so let us use theirs. On Tuesday, just after the market closed, Bank of America was penalized more than $1.2 billion by the United States government. Not as an accusation. As an admission. The bank agreed, in writing, that it had willfully failed to report more than $5 billion in suspicious transactions, over a billion of it tied by the Treasury to Chinese money-laundering networks and the southwest border. Willfully. The regulator chose the word and the bank signed beneath it. And then, because nothing in my profession is ever quite as damning as it first appears, a footnote explained what “willfully” means here. It means “reckless disregard or willful blindness.” Not that they meant to. That they arranged not to look. Which is, if you read the whole 64 pages, the single most honest sentence in the document. Bank of America reported its quarterly earnings the next morning. Hold those two facts side by side, because they are not a contradiction. They are a schedule. You do not drop a billion-dollar penalty the evening before earnings by accident. You do it the way you book a root canal for a day you were taking off anyway. One news cycle of pain, folded into one news cycle of profit, and by lunch the stock has decided it feels fine. The analysts had a phrase for it. They called it “clearing an overhang.” An overhang is what my industry calls a crime once it has been successfully priced. Here is the sentence they train you never to say, and I am long past the building, so I’ll say it. To a bank this size, a fine is not a punishment. A fine is a fee. The $1.2 billion is about two weeks of this bank’s profit. Two weeks. You do not deter a man by charging him a fortnight. You invoice him, and he pays, because his credit is, and I mean this with professional admiration, excellent. Now let me show you how the crime is actually committed, and here is the part that changed my life, so attend. I used to have to tell people how the dial worked. I would describe it and they would squint, because it sounded too cynical to be real. I don’t have to describe it anymore. The bank described it. Under oath. The machine watches transactions and raises its hand when one looks suspicious. Each raised hand is an alert; alerts are scored; a bundle of them reaches an actual human being only when the score crosses a threshold. And that threshold, the single number that decides how much crime the bank is permitted to notice, was, in the words of the order, “initially set by stakeholder judgment.” Stakeholder judgment. A man picked a number. Alerts older than thirteen months were “retired,” which is a beautiful word for a suspicious transaction the bank has decided to let die of old age. You tune the dial. And you do not tune it to the amount of crime. You tune it to the number of people you’ve hired to read the alerts. I can prove this now, which is new. In 2022 the bank ran its own test. It found that if it loosened the threshold, it would catch suspicious activity 85.4 percent of the time instead of 20.2 percent. Four times the crime, caught, by the bank’s own math. And it declined. In writing. The reason it gave, and I want you to hold this the way you’d hold something still warm, was that the additional cases “may exhaust the investigation resources.” Read that again. They did the test. The test said: you are catching one in five, and you could catch four in five. And the bank wrote down that catching four in five would be too much work. Not too expensive to be worth it. Too much work. The crime was findable. They measured exactly how findable. And they filed the measurement under reasons not to look. It gets better, by which I mean worse. A year later an internal test came back with a number somebody upstairs didn’t like. So, per the order, the Global Head of Financial Crimes “directed a senior manager to repeat the test.” They ran it again. The second time it produced, and this is the regulator’s phrasing, “a more favorable result,” after which the team declared itself “confident that EP is working as intended.” When the test told the truth, they ran it again until it lied. Then they wrote that the system was working as intended. It was. That was the problem. And they had a term, an actual line item, for the suspicious activity they chose to stop catching. When they recalibrated the dial in 2021 and cut eight thousand cases, the order notes the resulting loss of filed reports was called “SAR loss,” and deemed “permissible.” Two words, side by side, in a real document kept by a real bank: a loss, and permission. Nobody ever wrote down who it was permissible for. Let me give you the detail I think about on the lake, now, in retirement, when I am supposed to be fishing. A SAR, a suspicious activity report, is the thing a bank files when it has reason to suspect dirty money. For foreign banks, the ones most likely to be moving a cartel’s cash, you keep a due-diligence file. The file is the whole point. The file is the thing that is supposed to tell you whether you are helping launder the proceeds of murder. And the order records that these files “often consisted of” graphs. Of the bank’s own stock price. That was the file. A chart of us, getting richer, going up and to the right. Someone opened the folder that is meant to say is this foreign institution dangerous, and inside was a picture of our share price. The diligence was a mirror. I have never read anything truer about the industry, and I helped build it. You want to know how it ends up like this. It ends up like this in meetings, and the meetings are in the record too. Compliance flags a client, a hundred cases, overwhelming evidence of scam-victim money flowing in, and you can feel the officer’s human heart in the file, because he types, verbatim, “YESSSSSSSSSSSS.” Finally. We caught one. And the business line, the people who own the revenue, write back that “a closure here would be detrimental given the impact of this relationship.” And the recommendation to close the account is withdrawn, in favor of a “mitigation plan,” and the mitigation plan, I am not inventing this, includes asking the client to take its account number off its website. The client kept taking the victims’ money for another year. The bank’s stated corporate philosophy, printed on the walls, is “Responsible Growth.” The growth was responsible. To the revenue. That is the only constituency the word was ever pointed at. Now. You think this is new. It isn’t new. It’s only documented now, which is different, and I’d like to take you through the family album, because a profession should know its saints. At HSBC the Sinaloa cartel’s cash came through the teller windows in boxes. Not ordinary boxes. Boxes, per the record, “measured to slide neatly under the teller windows.” Sit with the carpentry of that. Someone measured the window. The laundering was so routine, so expected, so built into the furniture, that the criminals manufactured cash boxes to a tolerance spec, like a part. HSBC paid $1.9 billion, about five weeks of profit, and not one executive went to prison, and a United States Attorney General stood up in the Senate and admitted, out loud, that some banks had become “so large” that prosecuting them was “difficult,” because it might harm the economy. Too big to jail. Not my phrase. The government’s confession, offered freely, under the lights. A senator asked the obvious question at the time, and it has never been answered, so I’ll leave it with you the way she left it: how many billions do you have to launder for drug lords before someone shuts the bank down? And then she answered it herself. If you’re caught with an ounce of cocaine, you go to jail. Launder a billion for the men who sell it, and your company pays a fee and you go home and sleep in your own bed. At TD Bank, more recently, they left eighteen trillion dollars in transactions unmonitored, because of a budget rule they called, internally, the “flat cost paradigm,” which is the dial, promoted to a company value. And the launderers, who chose TD because, in one employee’s words, “we are convenient,” paid a bank employee for his help. Not in cash. In gift cards. Fifty-seven thousand dollars in gift cards. The price of an American bank’s conscience, it turns out, is a stack of plastic you redeem at the mall. And the whole apparatus, the $206 billion a year the world spends on compliance, the thousands of us with our titles and our attestations, catches, by the United Nations’ own estimate, about 0.2 percent of the money it exists to stop. One five-hundredth. I want to be clear that this is not a failure. A thing that performs at the identical level for decades is not failing. It is functioning. 0.2 percent is the spec. We hit it every year. We would catch more, but it may exhaust the investigation resources. So what was I, in all this? I had a title with the word “integrity” in it, and I have come to understand what that title is for. There is no Department of Honesty. You only name a role after the thing the building does not contain. I was the crumple zone. I was the part engineered, at great salary, to sit at the precise coordinate where the blame arrives, so that it stops at me and does not continue up the stairs to the people who set the revenue targets that required the dial. The blame comes down like water, past everyone who decided, and it pools at the lowest officer who signed something, and that officer was me, and we called it accountability and I called it a pension. Here is the number I cannot put down, the one the order gives you at the very end, after all the rest. When they were finally forced to fix the dial, to turn it back to where it should have been, it required seven hundred and fifty new staff, plus eight hundred consultants, and it doubled the monthly caseload. And FinCEN calculated that by not doing this, by keeping the dial where it was, the bank had “likely saved hundreds of millions of dollars.” So we know, now, exactly what it would have cost to see clearly. Seven hundred and fifty people. They existed. They were hireable. The bank could have afforded them the way you can afford lunch. The conscience was never too expensive. It was itemized, and compared against the fine, and the fine came out cheaper, and two weeks of profit is a very reasonable price for twenty years of not looking. This is the bank’s second order for the same failure. The last one, two years ago, carried no penalty, and in between the two the bank told its shareholders the whole thing would have “no material adverse financial impact.” The new one comes with a mandatory “lookback,” “independent reviews,” and the regulator’s explicitly reserved right to fine them again. That is not a punishment. A punishment ends. This is a subscription. It renews. The relationship between a bank this size and its regulator was never cop and criminal. It is landlord and tenant, and the fine is the rent, paid on a schedule, for the right to keep living in a building you are slowly, profitably, burning down. The Treasury Secretary said even the biggest institutions are not above the law. He’s right. They’re not above it. They’re downstream of it, which is a far better address. Above the law you’re exposed; the weather finds you. Downstream, you simply wait, and everything the law was built to stop flows gently down to you, having first been measured, scored, retired, rescored, risk-accepted, declared permissible, and cleared by Friday. I’m retired now. I have the pension, and the lake house, and a conscience I keep in a drawer with the attestations. The bank says it is committed to doing better. I believe it. It’s in the budget. Right next to the fine, which was cheaper.