of firms in this particular study was a wash, but the way they made money was different. Concentration of power brings more volatility with it, though this too can have upsides as well as downsides. In a paper published in 2001 Renée Adams of the University of Oxford and her co-authors found that the degree of centralisation in a firm (measured by, among other things, whether the chief executive was a founder and whether there was an independent chairman) was correlated with variability in corporate performance. The smaller the number of people exercising power inside a firm, in other words, the more likely it is that their decisions will have extreme consequences, for good and bad. Perhaps the truest test of founder mode comes after the founder has gone. That’s not just when it becomes clear whether there is any sort of succession plan; it’s also when it is revealed whether a firm has been developing a culture or a cult. An omnipotent founder can make companies dependent on them, or they can instil values and standards designed to survive their own tenure. Businesses like Apple and Disney still bear the imprint of their long- gone creators. What will Nvidia stand for without Mr Huang in charge?■ Step inside the world of work with our Bartleby newsletter. Each week our white-collar oracle muses on the agonies of office life. This article was downloaded by zlibrary from https://www.economist.com/business/2026/11/18/just-how-central-should-one-person- be-to-an-organisation

Business · Business | Schumpeter

Corporate spin-offs are getting out of control Some are enormously successful, but certainly not all Nov 19th 2026 No boss wishes they ran a smaller company. That’s why spin-offs, where a division of a firm becomes a free-standing enterprise, are considered victories of reason; of focus over sprawl; and of shareholders over managers. Spin-offs often take place under pressure from activist investors. Stock analysts delight in the simplicity of the cleaved companies, and hedge-fund traders in the complexity of the separation. Buying spun-out shares is a foolproof strategy, wrote one investor, Joel Greenblatt, in “You Can Be a Stock Market Genius”. The book is dedicated to his wife and their “three magnificent spin-offs”. The market is filled with companies that are said to illustrate the virtues of a big spin. GE Vernova, a descendant of the industrial conglomerate, has a years-long backlog of orders for its gas turbines. A second great shortage

caused by the artificial-intelligence build-out has vindicated a second great spin: Sandisk, which makes memory chips, was carved out last year. One half of the first big merger in the AI supply chain is a corporate division spun-good: Lumentum, which makes lasers and has agreed to combine with chip-designer Marvell, is worth about $50bn—seven times as much as its erstwhile parent. Thanks to such superstars, the spin-off index has heartily outperformed the market. To understand an age, examine its spin-offs. In the early 2020s few things were considered more toxic than skinny models and coal mining. Victoria’s Secret was liberated from Bath & Body Works, a soap-seller, in 2021. (The lingerie brand has thrived since.) Energy firms raced to sell their dirtiest assets, with Anglo American hiving off its South African coal (that, too, has excelled). Today firms are more likely to part with assets they hope the market will adopt as AI darlings. Flex, an American engineering firm, will soon carve out its data-centre business. The biggest listing in Hong Kong next year will probably be a spin-off of Alibaba’s chipmaking unit. But the very largest spin-offs are ones that allow bosses to reclaim past exuberance as present sobriety. Like clockwork, the biggest mergers of 2018 are becoming the biggest spin-offs of today. CVS, a health-care giant, plans to separate its pharmacy business in the next 18 months, undoing the giant merger it undertook less than a decade ago. One big media tie-up of 2018, Comcast’s acquisition of Sky, will soon be unwound by carving out the European broadcaster alongside NBCUniversal. The spin-off is always the first idea of a mogul who is short of them. In Hollywood, the result is such titles as “The Mandalorian and Grogu”, the latest Star Wars flick. In New York, it is the equally unappealing “Versant”, a collection of cable-TV networks including CNBC and the Golf Channel, which Comcast separated out earlier this year. Neither has received a particularly warm reception. Spin-offs take years. And like celebrity divorces they can be costly, unflattering and drawn-out affairs. Assets are hardly ever neatly divided. Whom to burden with a firm’s debt is often a contentious issue. So is management. Another 2018-vintage merger soon to be undone is that of Keurig, a coffee-giant, and Dr Pepper, which makes fizzy drinks. Last month the firm unveiled the new boss of “Global Coffee Co.”—the third announced chief executive of a company that doesn’t yet exist. One of the worst deals

of 2018 was the tie-up of Bayer, a German drugs giant, and Monsanto, an American crop-science business which has been sued over allegations that Roundup, a weedkiller it sells, causes cancer (it denies this). The recent ring- fencing of Bayer’s Roundup division has analysts speculating that it will lead to yet another mega-spin. If a spin-off fails there is a simple solution: more spin-offs. During the past decade the market has shown an endless capacity to disassemble industrial conglomerates until they either emerge as healthy businesses or become so small that investors stop caring. General Electric is the great poster child for spin-off success: its three divisions have done so well that the market has even allowed them to start buying other businesses again. Things have gone less smoothly at Honeywell, which was chopped into five publicly traded firms now worth a combined $140bn (little more than in 2018, before the break-up began). Shares in its aerospace unit have sunk 40% since they were listed in June. Resideo, a smaller, residual part of the conglomerate, was further split into two pieces in August—turning the Honeywell five into six. Solstice, another Honeywell alumnus, attempted, and swiftly abandoned, a merger this summer. Since then an activist pitched up demanding that the spin-off stick to its knitting. The experience of other ex-conglomerates is closer to Honeywell than to General Electric. The result is a dizzying state of infinite spin. You can imagine an industrial boss tied to a chair while shareholders slowly cut off limbs, shouting at him to “Focus!”. One heir to 3M, which was split up in 2024, said this summer that it also would become two. DowDuPont, which broke in three in 2019 in one of the most complicated deals ever, recently became five: Corteva, another chemicals business beset by lawsuits, shed its seed business in October. The great powers of European industry have been divided into even smaller territories with even stranger names. Siemens, a German one, has birthed both Siemens Healthineers and Siemens Energy (soon the latter will be re- christened “Omterra”). Both have recently announced further breakups of their own. As the spins become smaller, their names become more abstract. ThyssenKrupp, another shattered German, controls both the listed, lower- case “nucera”, which it spun out in 2023, and “accelis”, which went last

month. There’s no end to the complexity created in the name of corporate simplicity. ■ This article was downloaded by zlibrary from https://www.economist.com/business/2026/11/18/corporate-spin-offs-are-getting-out- of-control

· Finance & economics

America does not produce world-beating statistics. Who does? Why startups still go public, eventually Does Donald Trump care about the economy? Bond yields are rising everywhere—except China Meet the European Central Bank’s new squad Europe tightens sanctions on Russia. America loosens them AI is swallowing maths. Is economics next?

Finance & economics | Measuring the measurers America does not produce world-beating statistics. Who does? We search for the world’s worst—and best—economic data Nov 19th 2026 When King Ariantas wanted to find out how many people lived in ancient Scythia, his approach to people not answering surveys was brutally simple. According to Herodotus, a Greek historian, Ariantas ordered each Scythian to bring him an arrowhead on pain of death. As the king totted up his arrowheads, he could not be sure that everybody had complied, but woe betide those that did not. Modern-day statisticians would be forgiven for envying Ariantas’s powers of persuasion. Response rates to official surveys have collapsed globally. In America less than 40% of people reply to questions on consumer spending, down from more than 60% in 2016 (see chart 1). In Britain a paltry 27% reply to surveys about their work status. Bean-counters today also have to

reckon with falling budgets, declining public trust and greater political interference. Only half of Americans have confidence in federal numbers. President Donald Trump did not help by sacking the head of America’s Bureau of Labour Statistics (BLS) in 2025, falsely claiming she had rigged a jobs report to make him look bad. Statisticians in Colombia, Poland and Slovenia have all been controversially fired in recent years. Everyone should worry about the stats-pocalypse, not least because policy drawn up using junk statistics risks being junk itself. Central banks need reliable employment data to set interest rates. Many social benefits are linked to inflation, so if the numbers are wrong, payments will be stingier or more generous than intended. If American lawmakers had known in the last quarter of 2008 that GDP had fallen by more than 8%, at an annualised rate, rather than the initial estimate of 4%, they might have acted more boldly in the face of the global financial crisis. The rise of artificial intelligence makes this a dangerous time, since errors tend to get worse when the economy is changing fast. Luckily, some places still produce numbers worth trusting. The World Bank compiles an annual index of statistical performance, measuring basic things like how frequently numbers are produced and whether they are available online. The quality of a country’s statistical system is, predictably, tied to