Philippa Kaur

In licking county, Ohio, fleets of dump trucks and bulldozers are shifting earth on the future site of chip factories. Intel is building two “fabs” there at a cost of around $20bn. In March President Joe Biden called this expanse of dirt a “field of dreams” in his state-of-the-union speech. It was “the ground on which America’s future will be built”, he intoned.In the spring it was easy to be dreamy about America’s chip industry. The pandemic-induced semiconductor crunch had proved just how crucial chips were to modern life. Demand was still rising for all sorts of chip-powered technology, which these days is most of it. Investors were less gloomy on chips than on other tech, which was taking a stockmarket beating. The CHIPS act was making its way through Congress, promising to plough subsidies worth $52bn into the domestic industry, in order to reduce America’s reliance on foreign fabs and support projects like Intel’s Ohio factory.Half a year later the dreams look nightmarish. Demand for silicon appears to be falling as quickly as it had risen during the pandemic. In late September Micron, an Idaho-based maker of memory chips, reported a 20% year-on-year fall in quarterly sales. A week later AMD, a Californian chip designer, slashed its sales estimate for the third quarter by 16%. Within days Bloomberg reported that Intel plans to lay off thousands of staff, following a string of poor results that are likely to continue when it presents its latest quarterly report on October 27th. Since July a basket of America’s 30 or so biggest chip firms have cut revenue forecasts for the third quarter from $99bn to $88bn. So far this year more than $1.5trn has been wiped from the combined market value of American-listed semiconductor companies (see chart). The chip industry is notoriously cyclical at the best of times: the new capacity built in response to rising demand takes several years to materialise, by which time the demand is no longer white-hot. In America this cycle is now being turbocharged by the government. The chips act, which became law in August to cheers from chip bosses, is stimulating the supply side of the semiconductor business just as the Biden administration is stepping up efforts to stop American-made chips and chipmaking equipment from going to China, dampening demand for American products in the world’s biggest semiconductor market. Whether or not it makes strategic sense for America to bring more chip production home and to hamstring its geopolitical rival with export bans, the combination of more supply and less demand is a recipe for trouble. And if the American policies speed up China’s efforts to “resolutely win the battle in key core technologies”, as President Xi Jinping affirmed in a speech to the Communist Party congress on October 16th, they could give rise to powerful Chinese competitors. Field of dreams? It is enough to keep you awake in terror at night.The cyclical slump has so far been felt most acutely in consumer goods. PCs and smartphones account for almost half the $600bn-worth of chips sold annually. Having splurged during the pandemic, inflation-weary shoppers are buying fewer gadgets. Gartner, a research firm, expects smartphone sales to drop by 6% this year and those of pcs by 10%. Firms like Intel, which in February was telling investors it expected PC demand to grow steadily for the next five years, are revising their outlooks as it becomes clear that many covid-era purchases were simply brought forward. Many analysts think that other segments could be next. Panic buying amid last year’s global chip shortage has left many carmakers and manufacturers of business hardware with inventories overflowing with silicon. New Street Research, a firm of analysts, estimates that between April and June industrial firms’ stock of chips was about 40% above the historic level relative to sales. Inventories for pc-makers and car companies are similarly full. Intel and Micron blamed their recent weak results in part on high inventories. The supply glut and sputtering demand is already hitting prices. The cost of memory chips is down by two-fifths in the past year, according to Future Horizons, a research firm. The price of logic chips, which process data and are less commoditised than memory chips, is down by 3% in the same period Chip buyers will work through their inventories eventually. But after they do, they may buy less than in the past. In August Hewlett Packard Enterprise and Dell, two big hardware makers, hinted that demand from business customers was beginning to soften. Sales of both pcs and smartphones had started to plateau before the pandemic and this trend will probably resume in the coming years. Phonemakers cannot stuff ever more chips onto their devices for ever. For companies such as Qualcomm, which derives half its sales from smartphone chips, and Intel, which gets a similar share from those for pcs, that is a headache.The chipmakers’ response has been to bet on fast-growing new markets. amd, Intel and Nvidia, another big chip-designer, are battling over the cloud-computing data centres, where chip demand is still increasing. Qualcomm is diversifying into cars. In September the firm’s bosses boasted it already had $30bn-worth of orders from carmakers. Intel, meanwhile, is expanding into semiconductors for networking gear and devices for the hyperconnected future of the “internet of things”. It is also getting into the contract-manufacturing business, hoping to win market share from tsmc of Taiwan, the world’s biggest chipmaker and contract manufacturer of choice for fabless chip-designers such as amd and Nvidia.These efforts, however, are now running into geopolitics. Like their counterparts in China and Europe, politicians in America want to lessen their countries’ dependence on foreign chipmakers, in particular tsmc, which manufactures 90% of the world’s leading-edge chips. In response, America, China, the eu, Japan, South Korea and Taiwan together plan to subsidise domestic chipmaking to the tune of $85bn annually over the next three years, calculates Mark Lipacis of Jefferies, an investment bank. That would buy a fair bit of extra capacity globally. At the same time, prospects for offloading the resulting chips are darkening, especially for American firms, as a result of America’s tightening controls on exports to China. Many American firms count the Asian giant, which imported $400bn-worth of semiconductors last year, as their biggest market. Intel’s Chinese sales made up $21bn of its overall revenues of $79bn last year. Nvidia said that an earlier round of restrictions, which limited sales of advanced data-centre chips to Chinese customers and to Russia after its invasion of Ukraine, would cost it $400m in third-quarter sales, equivalent to 6% of its total revenues. The latest restrictions, which target Chinese supercomputing and artificial-intelligence efforts, are a particular concern for the companies which manufacture chipmaking tools. Three of the world’s five biggest such firms—Applied Materials, kla and Lam Research—are American. The share of the trio’s sales that go to China has risen fast in the past few years, to about a third. Toshiya Hari of Goldman Sachs, a bank, says that the controls are likely to cost the world’s toolmakers $6bn in lost revenues this year, equivalent to 9% of their projected sales. After the new American export controls were unveiled Applied Materials lowered its expected fourth-quarter revenue by 4% to $6.4bn. Its share price has fallen by 13% in the past two weeks. Those of kla and Lam More

It is night-time at the Soapstone Comedy Club. In fact, it always is. The club is a space in Horizon Worlds, Meta’s flagship metaverse app, where users can watch and perform comedy in virtual reality (vr). “It’s hard to do stand-up when you have no legs,” quips one performer, gesturing to his hovering avatar, before accidentally dropping the virtual microphone and floating offstage. A night out in vr lacks some of the atmosphere of a real bar, though it does cause authentic dizziness and nausea.It is almost a year since Mark Zuckerberg announced that his company would change its name from Facebook to Meta, to reflect its commitment to the metaverse and, no doubt, to escape the firm’s toxic public image. Many were unsure what the word meant, but with the company’s value at a near-all-time high of $1.1trn, and its core social-network advertising business humming away on the back of a pandemic boom, investors were willing to indulge the experiment.A year on, things look different. The metaverse on which so much has been staked remains unproven and unpopular. On October 16th the Wall Street Journal reported that, according to internal Meta documents, user numbers had declined since the spring. Meanwhile there are signs that both users and advertisers are drifting away from the social networks that pay Meta’s bills. Since its rebranding the company’s share price has dropped by 60%, destroying more than half a trillion dollars of market value (see chart 1). Forecasts for profits in 2023 have fallen by about 50%, according to data from Bloomberg. Meta’s next earnings results, due on October 26th, represent an “existential quarter”, says Mark Shmulik of Bernstein, a broker.What has gone wrong? The sell-off of Meta stock began in February, after the company reported its first-ever drop in daily users of Facebook, its first and largest social network. After 18 years of uninterrupted growth it lost 1m of them between January and March (see chart 2). It has since bounced back, adding 39m more, while users of Meta’s “family of apps”, which includes Instagram and WhatsApp, have kept growing. But the new users increasingly come from poor countries, and are therefore less valuable to advertisers. Last year Frances Haugen, a whistleblowing former Meta executive, claimed that in Facebook’s five most valuable markets, account registrations for under-18s had fallen by a quarter within a year. Meta has hurried out a new short-video product, Reels, to stem the bleeding to TikTok and other new rivals.As users wobble, so do advertisers. In the second quarter Meta’s revenue fell year on year, for the first time in its history (see chart 3). Inflation, interest rates and war all played a part. But the ad business has been permanently changed by Apple’s new rules. These make it harder for apps to track users’ online activity, which in turn makes it harder to serve them relevant ads and see whether they work. Meta has said that Apple’s changes will cost it $10bn this year in forgone revenue. Companies are shifting their advertising to what admen call the bottom of the funnel: points at which the consumer is close to a purchase (Amazon, which serves ads to customers based on what they have just searched for, has been a big beneficiary).Meta is better equipped than many of its rivals to overcome these obstacles. Reels already accounts for more than 20% of time spent on Instagram, and is making more money than Instagram’s successful Stories feature did at the same stage of its introduction, the company says. Heavy investment in artificial intelligence (ai) is helping Meta develop “probabilistic” ad models to replace the signal that was lost with Apple’s changes. Advantage+, a recent Meta ad product, uses ai to help advertisers develop and place ads. A trickier ad business serves to widen Meta’s competitive moat, points out Mr Shmulik: smaller rivals like Snap, whose share price has fallen by nearly 90% in the past 12 months, are the real casualties. Still, Meta’s advertising franchise has probably been permanently impaired. And the company is scrambling to rebuild its ad business without the architect of its previous one, Sheryl Sandberg, who left the company last month.All this would be enough to give investors jitters. The fact that Meta is simultaneously making a colossal bet on the metaverse threatens to test their faith to breaking point. Reality Labs, the company’s metaverse division, has so far run up losses of $27bn. Meta has sold more than 17m Quest 2 vr headsets, estimates idc, a data company, mostly at or below cost. It has also been on a hiring spree, last year announcing 10,000 new metaverse jobs in Europe. The pace of hardware development continues: on October 11th the company unveiled a more advanced Quest Pro headset, and Mr Zuckerberg showed off prototype hardware including a wrist-worn neural-input device. A Quest 3 and Quest Pro 2 are already in the works.When—or whether—the metaverse will take off remains unclear. The Quest’s main use so far is gaming. Fitness is a growing niche, though Meta’s attempt to buy Within, a maker of vr fitness apps, has been blocked by antitrust regulators. The Quest Pro is aimed at businesses; on its launch this month Meta announced a partnership with Microsoft, which will provide vr versions of apps like Teams and Office. A “Quest for Business” subscription will be available next year. But the social uses of vr, about which Mr Zuckerberg is most enthusiastic and where Meta should have the greatest advantage, remain unpopular. In February Meta reported that just 300,000 people had used Horizon Worlds; the firm has said nothing since. A leaked internal memo suggested that even company employees were having to be cajoled to use it (“If we don’t love it, how can we expect our users to love it?”). Mr Zuckerberg is hardly the only one who sees potential in vr. In the first half of next year Apple is expected to release its debut headset, and Sony will launch its latest gaming-focused goggles for its PlayStation console. If headsets do become the new pcs, as Mr Zuckerberg has predicted, Meta will have a considerable first-mover advantage. The Quest 2 accounted for 88% of global vr headset sales in the first half of this year, says idc. The Quest Pro is the most advanced set of vr glasses around. Meta’s hiring binge means it has much of the top vr talent, says Jitesh Ubrani of idc. If Meta can control and tax a successful vr platform, as Apple and Google control their smartphone operating systems, it will own a gold mine (Meta already skims off as much as 47.5% from Horizon Worlds purchases).The question is timing. Meta’s unusual structure gives Mr Zuckerberg total control. The firm’s board proved to be ineffective at dealing with Facebook’s scandals over privacy and misinformation. Now, rather than urge caution, it has allowed a flawed chief executive to gamble billions on the metaverse. In May Mr Zuckerberg admitted as much when he told Protocol, a news site: “If people invest in our company, we want to be profitable for them…But I also feel a responsibility to go for it…[Meta] is a controlled company, so I can make more of these decisions than most companies would.” Yet the more Meta’s core business wobbles, the less investors will be willing to give Mr Zuckerberg’s metaverse plans the benefit of the doubt. A company can only spend that much on a new idea if investors are willing to fund it. They might be willing if “your core profitability from your core business is on solid footing”, says Mr Shmulik. That is Meta’s difficulty. “The core isn’t on a solid footing at the moment.”To calm investors’ nerves, Meta is reining in its spending a little. It expects its total expenses this year to be about $7.5bn lower than it forecast at the end of last year. It has scrapped some projects, including a smart watch that was in development, and bumped up the price of the Quest 2 by $100. And it expects to reduce its headcount.Meta executives compare the company’s predicament now to ten years ago, when it was managing the transition of its social network to mobile. Shifting a billion Facebook users from desktop to phone was no mean feat, made harder by the fact that Mr Zuckerberg was late to spot the importance of mobile. That experience may have influenced his approach to the metaverse. Meta’s new vr technology, he said on October 11th, was for those “who’d rather be early than fashionably late”. The risk, as investors grow impatient, is that this time Meta has made its move too soon. ■ More

ASKED ABOUT the state of the economy, Americans are surprisingly gloomy. More than half say they are experiencing financial hardship; more than a third say they are having difficulty paying for regular household expenses. Yet, even as surveys suggest that Americans are tightening their belts amid persistently high inflation, data show that they continue to spend at a healthy clip. Last month the Bureau of Economic Analysis reported that consumer spending is growing by 1.8% year on year after adjusting for inflation—not far from its historic average. A report by the Bank of America Institute, a think-tank, finds that consumer payments are growing at double digits, a sign that the American shopper “is still spending”. This resilience can be explained in part by mattress-loads of savings. Americans accumulated more than $2trn in excess savings during the pandemic, when the federal government doled out unemployment benefits and stimulus cheques even as households cut back on travel, entertainment and eating out. Although some of this has been spent, households are still sitting on a $1.4trn cushion, reckons Ian Shepherdson of Pantheon Macroeconomics, a consultancy. The labour market is healthy, too. Unemployment has fallen to 3.5%, the lowest it has been in 50 years. In August there were 10.1m job openings, or 1.7 vacancies for every jobless person. But another less-appreciated reason why spending has been so steady in the face of soaring inflation is a shift in consumers’ sensitivity to prices, or “price elasticity of demand”. This concept, seldom mentioned outside economics textbooks, has been a hot topic of debate among investors and company executives in the past year (see chart 1). The term has found its way to the earnings calls of consumer-goods giants such as PepsiCo, whose bosses talked of favourable “demand-elasticity trends” while presenting the food-and-drinks giant’s unexpectedly bubbly quarterly results on October 12th. The available data appear to back them up. Figures compiled by IRI, a market-research firm, suggest that consumers are indeed significantly less price sensitive now than they were before the pandemic. Using scanner data on prices and sales recorded with each purchase of thousands of items across more than 125,000 supermarkets, chemists, dollar stores and big-box retailers, IRI estimates that price elasticities have fallen for 22 out of 25 product categories since February 2020, and remained flat for the other three (see chart 2). All told, IRI reckons that consumers were roughly 20% less price sensitive in the 52-week period ending September 4th than they had been in the year before the pandemic. Why the shift? Experts offer three possible reasons. First, as panic-buying led to empty supermarket shelves in the early months of the pandemic, consumers adjusted their shopping routines and tried brands they weren’t used to, says Brett Gordon, a marketing professor at Northwestern University. With more time at home, people also became more comfortable splurging on pricier food and household items. Last, consumers cut the time they spent shopping—by roughly 9% between 2019 and 2021 according to government statistics. The way they use that has changed, too. “A lot of people maybe spent more time shopping for things to outfit their homes, but less time worrying about everyday consumer products,” says Alexander MacKay of Harvard Business School. There are some signs that consumers are starting to pull back. Walmart, a retailing behemoth, says that its shoppers are switching from pricey deli meats to hot dogs, and from gallons (3.8 litres) of milk to half-gallons. Best Buy, an electronics retailer, says its customers are increasingly opting for private-label tvs over name-brand sets. Such shifts in consumer behaviour are most pronounced among lower-income households. tjx, a discount department store, says that, for the first time in years, outlets in higher-income areas are growing faster than those in lower-income ones. “Middle-income and high-income consumers are continuing to spend,” explains Krishnakumar Davey of IRI, but “low-income stores and low-income consumers are pulling back a little bit.” This will be on the minds of investors as America’s listed companies report their quarterly earnings in the coming weeks. Those hoping for clear answers may be disappointed. Although packaged-goods firms agree that shoppers will start to balk at higher prices, there is far less consensus about when exactly this will happen. As James Quincey, boss of Coca-Cola, told investors earlier this year, “I expect elasticity to increase at some point in the future. Will that be next quarter? Or will that be next year? I can’t give you the answer to that.” ■To stay on top of the biggest stories in business and technology, sign up to the Bottom Line, our weekly subscriber-only newsletter. More

The failure of the first serious attempt by China’s carmakers to conquer European markets, around 15 years ago, was self-inflicted. Their cars were terrible. The shabby quality of Brilliance’s “bs” range (no joke) was matched with looks that scarcely merited the word “design”. Since then the Chinese car industry has become the world’s biggest and its products have improved immeasurably. It churns out more electric vehicles (evs) than any other country, and many are anything but bs. It is also an ev-battery superpower. Listen to this story. Enjoy more audio and podcasts on More

Weeks after explosions caused leaks from Nord Stream 1 and Nord Stream 2, two undersea gas pipelines linking Russia and Germany, another act of suspected sabotage rocked Europe’s biggest economy. On October 8th Deutsche Bahn (db), the state-owned rail giant, said it needed to suspend all services in northern Germany for around three hours. Damage to cables indispensable for rail traffic had led to a breakdown of its wireless communication system. The incident left thousands of passengers stranded on a Saturday morning.Listen to this story. Enjoy more audio and podcasts on More

If you have a high temperature or are recovering from heart surgery, it is difficult to use machine tools. And if you are having a nervous breakdown, machine tools are best avoided. Sick days are the remedy. They are meant to prevent people from hurting themselves, their co-workers, customers or passers-by on the job. Working from home has flipped this logic on its head. If you can work from the kitchen table, today’s hybrid workers increasingly conclude, then why not from bed—so long as the brain is on and the Zoom camera off? Listen to this story. Enjoy more audio and podcasts on More

Visions of a technologically ascendent China keep American strategists up at night. They see the contours of a surveillance state implementing the will of President Xi Jinping by algorithmic edict at home and projecting computing power abroad. To erase those contours for good, on October 7th President Joe Biden’s administration announced the most sweeping set of export controls in decades. The new rules cut off people and firms in China from many advanced technologies of American origin, and from products made using these. The list includes chips used for artificial intelligence (ai), software to design advanced chips and the machine tools to manufacture them. Selling such things to China is now barred without explicit permission from America’s government. Rulebreakers risk being cut off from American tech themselves. The share prices of affected Chinese firms have sunk (see chart). China’s biggest producer of memory chips, the state-owned YMTC, has 60 days to allow American officials to inspect its operations for compliance. American companies that sell advanced semiconductor technology to China have also been hit, even as they reel from a deep cyclical slump in demand for their wares. This week it emerged that Intel, America’s chipmaking champion with Chinese sales of $21bn last year, is about to axe thousands of jobs. America has previously used similar rules to kneecap Huawei, China’s telecoms-gear giant. Jake Sullivan, Mr Biden’s national security adviser, boasted recently that export controls have forced Russia to “use chips from dishwashers in its military equipment”, which will “over time degrade [its] battlefield capabilities”. In the case of China, America’s goal is likewise no longer just to stay ahead of its rival in the tech race but to “put the high-end Chinese chip-design industry out of business”, says Greg Allen, a former defence-department official who has studied the new rules. Whether America gets its way depends on several factors. There are “real questions” about the rules’ legality, says Peter Lichtenbaum of Covington & Burling, a law firm in Washington. He expects someone to test the restrictions in court. Donald Trump’s administration was successfully sued over an executive order banning TikTok. Even legal export controls are leaky. Plugging the leaks requires more resources for the enforcers at the Commerce Department. “Their to-do list has exploded,” says Mr Allen. “Their budget has not.” And China imports $400bn-worth of chips a year, more than any other country. Though private companies and allied countries might be happy to go along with the Americans now, the amount of money being left on the table by not selling to Chinese customers may start to rankle. ■To stay on top of the biggest stories in business and technology, sign up to the Bottom Line, our weekly subscriber-only newsletter. More

Unlike tolstoy’s description of families, mergers and acquisitions that end happily do so for a variety of reasons. It’s the unhappy ones that are alike. This is particularly true of m&a deals done at the top of the business cycle, when hubris runs amok, lofty valuations make acquirers sloppy with their money and the most radical ideas are made to sound plausible. In this category sits Elon Musk’s shotgun wedding to Twitter, once again in the offing after a judge gave both sides until October 28th to consummate it. Mr Musk’s latest attempt to justify it is to describe it as a step towards a Chinese-style “everything app”. It is just as likely to go down in history as a top-of-the-market “deal from hell”. The annals of business have colourful examples of such Stygian mishaps. Sony’s ill-fated acquisition of Columbia Pictures in 1989 occurred when Japan’s bosses thought they were invincible, the bubble economy made any price appear worth paying, and dreams of the convergence of hardware (consumer gadgets) and software (entertainment) were in the air. AOL’s merger with Time Warner, an even bigger mess, was first announced in 2000 at the apogee of dotcom frothiness. The bosses of both companies, one an internet upstart, the other a fading media giant, fantasised about creating a colossus of the internet age. They torched nearly $200bn of value in a matter of months. In 2007 Royal Bank of Scotland (RBS), an acquisitive financial institution, led a consortium to buy ABN AMRO, a sprawling Dutch banking group. It was the biggest banking takeover in history—yet done with little due diligence or oversight of gung-ho executives, even as the world was on the brink of the great recession. It occurred shortly before RBS’s spectacular demise and a bail-out from the British taxpayer. Mr Musk’s approach to Twitter is different from these in one important respect. He is acting in a personal capacity as the world’s richest man. He has no known plans to integrate the social-media platform with Tesla and SpaceX, his electric-vehicle and rocket firms. Mercifully. Yet the stock phrases that sum up such debacles—wrong target, wrong time, wrong price tag—already seem applicable to his pursuit of Twitter, and may explain why he has spent so long trying to wriggle out of the deal. If the two sides do not reach an agreement later this month, the judge says she will haul them back to the Delaware Court of Chancery and decide their fate for them. Whatever the outcome, Robert Bruner, a professor of business at the University of Virginia who in 2005 wrote a book called “Deals from Hell” to explain m&a fiascoes, says Mr Musk’s Twitter saga already bears many subtler hallmarks of the genre. In Mr Bruner’s diagnosis, the first hints of hell come from hubris. The self-styled “Technoking” has every reason for self-belief. Tesla is the world’s most valuable carmaker. SpaceX is literally rocket science in action. Yet for executives like him it’s a fine line from that to overconfidence. Sony’s Morita Akio crossed it. So did AOL’s Steve Case and RBS’s Fred Goodwin. In Mr Musk’s case, excessive faith in his ability to turn Twitter around is exacerbated by a saviour complex: his main goal, he said when he announced the deal in April, was furthering the cause of free speech. That appears to have blinded him to the need for due diligence. Moreover, like other exalted leaders, he is surrounded by yes-men. Billionaires compete to throw money at him. No chairman of any board appears to put a restraining hand on his shoulder. For now his reputation for walking on water continues to sustain him. But if he has overplayed his hand, history will not let him off lightly. Just ask Messrs Case and Goodwin (Morita passed away in 1999). The corollary of hubris is sloppy financing, another attribute of top-of-the-market megaflops. This is particularly true at the tail end of bull markets, such as the one recently vanished in a puff of smoke. Not only was Mr Musk so unconcerned about overpayment that he based his $54.20-a-share offer for Twitter on an overused cannabis joke. Big banks jostled to back one of the world’s largest-ever buy-outs, even though by then cracks had started to appear in leveraged-loan markets. Yet as with many M&A deals, deteriorating markets can turn a flawed acquisition into a disaster. That possibility must haunt Mr Musk. The digital-advertising market on which Twitter depends has crumbled. Tesla’s own shares, the source of most of his wealth, have lost a third of their value since he made the bid (don’t cry for him, he is still worth $220bn). The deal financing includes $13bn of high-risk debt and spreads on this kind of instrument have soared. Whether Mr Musk reaches a deal with Twitter or the judge forces the sale to go ahead, the repercussions are likely to be troubling. Either banks are stuck with hard-to-sell debt and suffer hefty losses or, in the unlikely event that they abandon the deal, a superhero of 21st-century capitalism faces a $44bn day of reckoning.The X-factor Finally there is strategy. In Mr Bruner’s analysis, the worst M&A deals are done when the target is in an industry far beyond the acquirer’s “domain knowledge”. That is surely true of Mr Musk and Twitter. That may explain why he has started to offer hints of a grander strategic vision. He has raised the prospect of reducing Twitter’s reliance on advertising, and instead incorporating it into an “everything app”, known as X, with online payments that hark back to the days when he helped found PayPal. It is a tantalising idea. The model is WeChat, Tencent’s superapp in China. Others, like Meta, have tried it with mixed results. If it works, it would provide yet further testimony to Mr Musk’s ineffable genius. But it also has a hellish side. It could pit the world’s most powerful businessman against tech regulators. It could stir up trouble geopolitically (imagine a reinstated Donald Trump weighing in, as Mr Musk has done, on Russia and Ukraine). And it could enrage China, thwarting Tesla’s prospects there. Another deal for the history books, no doubt. ■ More