This week’s reading list is about the difference between a law of nature and a decision that has been left alone long enough to look like one. Global imbalances are debated at G7 and G20 level as though they were weather, when they are simply every country’s domestic saving and investment choices added together. The Strait of Hormuz is treated as a fact of geography, when what actually binds is the absence of an agreement specific enough to be verified. Europe’s ban on paying interest on electronic money turns out to have no economic foundation at all, only a rule nobody has revisited. Mosquito-borne disease in the United States is now a regulatory choice rather than a technical limit, the technology having sat in limbo for fifteen years. Companies scale before they build a business model because the people funding them are rewarded when they do, not because business requires it. And the one item on this list that genuinely is a law of nature, the distribution of how often companies actually compound at 20 per cent for a decade, is the thing investors most reliably treat as optional. The useful lesson is that it is worth asking, of anything that looks permanent, who decided this and when. The best investors are not only forecasting outcomes. They are working out which constraints are real and which ones are simply unexamined.

📚 Imbalances are a feature of the global financial system, not a bug: David Ronicle at Chatham House on why this year’s revival of the global imbalances debate, a flagship G7 report and a G20 finance ministers’ summit that could not agree a communiqué, is aimed at the wrong target. The numbers are certainly arresting. America’s net liability to the world reached about 70 per cent of GDP in 2025, a level last seen just before 2008. His argument is that imbalances are not a global phenomenon at all, just the sum of domestic saving and investment decisions, and that they function as a discipline rather than a disease, because foreign capital is mobile and will leave when policy is bad. He is also good on why the US position is more robust than it looks: American liabilities are in dollars while its assets are in foreign currency, so a rebalancing away from the US weakens the dollar and raises the value of those assets. The sharp line is that the imbalances are an early warning signal about domestic choices, and that treating them as a foreign problem is a way of avoiding the reforms each country already knows it needs. (Chatham House)

📚 Escaping the Hormuz trap: What should an agreement to reopen the Strait look like? Published on 30 September, which makes it the freshest thing on the list and the most directly relevant to anyone holding energy exposure. The argument is that reopening the Strait is not a matter of declarations but of designing a process detailed enough, reciprocal enough and verifiable enough that it actually holds, which is a far higher bar than the headlines around any ceasefire will suggest. Worth reading alongside the companion piece from 25 September on the pipelines and transport corridors that Iraq, Syria and Lebanon are now planning to route around the chokepoint, because that is the version of this story that persists long after the shooting stops. For investors, the useful distinction is between a crude price that responds to news and a regional logistics map that is being permanently redrawn. (Chatham House)

📚 Europe should stop banning interest on digital money: Ulrich Bindseil at Bruegel making a technical argument with larger consequences than its title suggests. The European Union prohibits paying interest on e-money and on stablecoins, and the digital euro is being designed the same way. His case is that non-remuneration functions as a distortionary tax, that it lacks any real economic justification, that it falls hardest on less wealthy households who hold more of their money in these forms, and that it is increasingly being circumvented anyway. Given how much capital is now moving into stablecoins and tokenised cash, the question of who is permitted to pay interest on digital balances is quietly one of the more consequential open questions in payments, and almost nobody outside the field is watching it. (Bruegel)

📚 The Scaling and Profitability Trade off: Venture Capital’s Weakest Link: Aswath Damodaran on why so many large private companies arrive at public markets without a business model, and why that is a design feature of the system rather than an accident. The data he assembles is the part worth keeping. More than 80 per cent of companies going public in the 1980s were profitable; less than a quarter of those listing in the past decade have been. Median age at listing has risen by about eleven years over the last fifteen, revenues at the median listing have tripled or quadrupled in real terms, and median market capitalisation at listing has exceeded a billion dollars over the last six years. His explanation is incentives: venture capitalists price companies rather than value them, are judged on entry price versus exit price rather than on the quality of the business built, and in the 2023 to 2026 period took 80 per cent of their returns from the top 1 per cent of investments. When the payoff is that concentrated, chasing scale is rational for the funder even when it is value destructive for the business. (Musings on Markets)

📚 Overpaying for Greatness: A Base-Rate Check: Vishal Khandelwal with the most useful piece on the list, opening with the Navi Mumbai airport, proposed in 1997 and finally opened on Christmas Day 2025, roughly seven years after the foundation stone and six after the first official deadline. The investing application is Michael Mauboussin’s base rate work. Across the thousand largest global companies by market value from 1950 to 2014, only about 4 in 100 grew sales at 20 per cent or more a year over the following decade. The median was 4.7 per cent. Nearly one in five finished the decade with lower real sales, and about half of listed companies do not survive ten years at all, so even that is a survivor’s table. Size matters enormously: 18 in 100 managed it from a base under US$325 million of sales, and none at all from US$50 billion. The arithmetic at the end is where it bites. Pay 70 times earnings, get your 20 per cent compounding for a full decade, accept a de-rating to 30 times, and you earn about 10 per cent a year. At 15 per cent growth you earn under 6 per cent, and at 12 per cent you earn about 3 per cent, having owned a wonderful company the entire time. (Safal Niveshak)

📚 Mosquitoes are a choice: Maya Rosen on the fact that the technology to eliminate several mosquito-borne diseases already exists and has spent roughly fifteen years in regulatory limbo. The specific approach she focuses on, seeding wild mosquito populations with Wolbachia bacteria so that they become resistant to viruses like dengue and pass that resistance on, involves no genetic modification at all, which is what makes the regulatory delay so hard to defend. The broader argument is the one that travels: biotech capability has advanced enormously while the rules governing deployment have not moved, so the binding constraint on a whole category of outcomes is now administrative rather than scientific. Anyone trying to work out where value accrues in life sciences should be thinking about approval pathways at least as hard as about the science. (Works in Progress)

📚 Why Alberta doesn’t have rats: Deena Mousa on a genuinely odd fact. Rats have colonised very nearly every place on earth where humans live, and one Canadian province is an exception, not by luck or geography but because of a sustained and deliberate programme. A good closing note for a week about decisions that get mistaken for destiny, and a reminder that the occasional unglamorous administrative effort does in fact work. No portfolio application whatsoever. We include it anyway. (Works in Progress)