
Why economists should like booze
Elites seem especially likely to snub the bottle. Three of the past four American presidents are teetotal (Barack Obama enjoyed a martini). In Silicon Valley temperance is a status symbol. Marc Andreessen, an investor, quit alcohol in 2022. Sam Altman of OpenAI writes about 'how much changed when people stopped drinking alcohol all day". Elon Musk refers to alcohol as a 'legacy drug". Dinner meetings with founders are fuelled by green tea.
An individual who gives up drinking can look forward to health benefits. They may lose weight. They may sleep better. Yet from an economist's point of view, teetotalism is an incoherent and damaging ideology—for three big reasons.
First, teetotallers are free-riders. For generations alcohol consumption has sustained all manner of social and economic structures. The abstemious benefit from them but do not contribute. For instance, non-drinkers who go to social events are free-riding on the joviality of hard-working drinkers. What would happen to the social fabric if everyone stopped imbibing? Perhaps Joe Strummer of the Clash, an English rock band, was on to something when he apocryphally said that 'non-smokers should be banned from buying any product a smoker created".
Or consider the economics of the restaurant industry. Alcohol offers higher profit margins than food as it requires less labour to prepare. Indeed, using official American data, your columnist estimates that booze accounts for all the profits of the restaurant industry. Drinkers subsidise non-drinkers. Those who order sparkling water can feel sanctimonious in the short run. But if no one orders a bottle of Bordeaux, many restaurants will go under. In San Francisco, Sobriety Central, they are closing by the dozen.
Second, abstinence makes people lonelier. For centuries alcohol has served a social function. It helps people relax. Taking a drink also signals to others that you are happy to be slower and more vulnerable—that you have left your weapon at the door—which puts them at ease. A study from 2012 in Psychological Science found that alcohol increases social bonding. Robin Dunbar of Oxford University and colleagues find that frequenting a pub improves how engaged people feel with their community, in turn raising life satisfaction. It is not a stretch to say that alcohol has played a big evolutionary role in fostering human connection.
Many couples credit alcohol, at least in part, for bringing them together. So it may not be a coincidence that the alcohol-shunning young are lonely. Americans aged 15 to 24 spend a third less time socialising than they did in the early 2000s. A study published in 2021 by Jean Twenge of San Diego State University and colleagues found 'worldwide increases in adolescent loneliness". Young people are having less sex than older generations. When it is harder to relax, partnering up is more difficult.
The third factor in favour of booze relates to innovation. Today the world sees fewer breakthroughs. Hollywood sustains itself on remakes or sequels, not originals. A recent blog by Peter Ruppert, a consultant, finds the same trend for music: 'the pace of genuine sonic innovation has slowed dramatically". A paper published in 2020 by Nicholas Bloom of Stanford University and colleagues concludes that new ideas are 'harder to find". Productivity growth across the world is weak. Something has gone terribly wrong in the way that Western societies generate new ideas.
In the short term, avoiding alcohol is helpful for working efficiently. If you have a big presentation tomorrow, it is a good idea to stay off the sauce tonight. But consider the kind of world that alcohol allows to exist—even if messily, unreliably and at some cost—and abstention seems less sensible.
For centuries creative folk, from Aeschylus to Coleridge to Dickens, have relied on alcohol for inspiration. In the 1960s, when productivity was soaring, everyone was drunk all the time. No other drug has played such a consistent role in human innovation. Being intoxicated opens up the possibility of accidents of insight. Purely rational, linear minds have fewer of the flashes of brilliance that can turn an art form or an industry upside-down. It allows brains to disconnect. A study of American painters in 1946 by Ann Roe of Yale University noted that 'a nightly cocktail before dinner may contribute to the avoidance of a state of chronic tension, especially...when creative activity is at its height."
Studies suggest that alcohol, deployed judiciously, can aid the creative process. Andrew Jarosz of Mississippi State University and colleagues have found that intoxicated people solved problems faster and 'were more likely to perceive their solutions as the result of a sudden insight". A similar paper by Mathias Benedek of the University of Graz, in Austria, and colleagues concludes that 'certain aspects of creative cognition benefit from mild attenuations of cognitive control". In the short run, intoxication may limit your brain's processing power—and that can be frustrating. The long-term effects are much less clear.
Call me old-fashioned
The best approach, as with most things in life, is moderation: not Ernest Hemingway-levels of drinking, but not abstention either. What leads to successful human relationships and breakthrough innovations remains poorly understood. So, even if you are a Silicon Valley whizzkid who wants to change the world, it is best not to mess around with traditions too much. Gin from the freezer, good vermouth, and a twist.
Subscribers to The Economist can sign up to our Opinion newsletter, which brings together the best of our leaders, columns, guest essays and reader correspondence.
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Mint
12 minutes ago
- Mint
With billions at risk, Nvidia CEO buys his way out of the trade battle
Jensen Huang, chief executive of California-based chip designer Nvidia, worked for months behind the scenes in Washington and Beijing to protect tens of billions of dollars in future sales from the heated U.S.-China trade rivalry. Huang told President Trump that restrictions on U.S. chip sales to China would backfire by pushing Chinese technology champions to achieve self-reliance. He advised the president to keep China hooked on American tech. As a sweetener, Huang said the company would invest as much $500 billion in the U.S. Huang's argument, along with the half-trillion-dollar offer from the world's most valuable company, appeared to seal the deal. The Trump administration decided last month to allow China to buy Nvidia's H20 artificial-intelligence chip, a surprising reversal that came shortly after Huang met with Trump. Nvidia had developed the H20 to comply with past export restrictions as a less powerful chip specially designed for China. The news sent Nvidia's stock up 4%, pushing its market capitalization further above the record $4 trillion mark. Beijing reciprocated by allowing a $35 billion deal involving U.S. chip-software makers that it had held up for about a year. In a previously unreported development, Chinese officials also froze an inquiry into an already-completed Nvidia deal. With both moves, China's leaders hoped Huang would keep lobbying Washington for loosened export controls. There was one last hitch. At a meeting with Huang in the White House last week, Trump made one more demand—that Nvidia give the federal government 20% of its chip sales to China in exchange for issuing the export licenses. 'If I'm going to do that, I want you to pay us something," Trump said, recounting the exchange at a news conference Monday. The unusual pay-to-play proposal, which hadn't been vetted by White House tech policy staff before Trump offered it, is expected to face legal and security questions. Huang, facing a choice of paying for long-term access to a market vital to his company or walking away, countered with 15%. He charmed Trump at the meeting, drawing the president a diagram showing how tariffs would be counterproductive for his goal of increasing U.S. chip production and signed it. Shortly after the meeting, Trump said he would exempt tech companies that invest in the U.S. from roughly 100% tariffs on semiconductor imports. The Trump administration's decision to allow China access to the H20 chip appears a watershed moment in U.S. policy. In Trump's first term, as well as through President Biden's, the prevailing trend was generally tougher control on tech sales to China. National-security concerns often received greater weight in the past than business interests. Many attribute the potential shift to Huang's lobbying effort. Nvidia CEO Jensen Huang greeting President Trump this year in Washington. The CEO has lavished praise on Trump and China, and the admiration is mutual. On a July visit to Beijing, Huang hailed Chinese technological progress. Privately, a lieutenant to Chinese leader Xi Jinping told Huang that Beijing appreciated how Nvidia was contributing to the world's AI advances. At a Washington AI summit in late July, Trump said he had looked at breaking up Nvidia because of its dominant market share. But that was before he got to know Huang and the industry. 'What a job you've done," Trump told Huang, who was seated in the front row, surrounded by administration officials and industry executives. 'America's unique advantage that no country could possibly have is President Trump," Huang said at the summit. As Elon Musk learned, winners in the Washington influence competition attract opponents. Lawmakers in both parties have challenged Huang, saying that if China gets Nvidia chips, China's AI companies could benefit, potentially aiding the Chinese military. 'These criticisms are misguided and inconsistent with the administration's AI action plan," an Nvidia spokesman said. 'The H20 will not enhance anyone's military capabilities but will help America win the support of developers worldwide." This account is based on interviews with current and former officials in both countries, people close to and working for Nvidia, as well as those familiar with Huang and his lobbying efforts. 'President Trump regularly talks to business leaders like Jensen Huang who are making historic investments in American manufacturing," White House spokesman Kush Desai said. 'The only special interest guiding the president's decision-making is the best interest of the American people." For years, Huang avoided the rough-and-tumble world of Washington politics, preferring to schmooze with the videogamers who were the original customers for his chips. The 62-year-old billionaire, a libertarian at heart, resisted the lobbying game, delegating the task to his lieutenants. As Nvidia's chips became one of the world's most strategic commodities because of their role in AI, Washington remained a no-man's-land for the Silicon Valley company. Huang didn't visit the White House or engage lawmakers meaningfully until 2023, when he joined the chief executives of Intel and Qualcomm to lobby against new curbs on tech sales to China by the Biden administration. Restrictions on chip exports to many countries friendly with the U.S. announced in the final days of Biden's term escalated the threat. When Trump won a second term, Huang faced an administration stacked with China hawks, and the prospect of even tighter U.S. controls on tech exports. The company bashed the Biden administration's restrictions and warned the rules would benefit such Chinese competitors as Nvidia-rival Huawei. Nvidia hired former Trump national security adviser Robert O'Brien, co-founder of a consulting firm. Robert O'Brien, former Trump national security adviser, during an interview at the 2024 Republican National Convention in Milwaukee. O'Brien helped communicate a sense of urgency regarding breakthroughs from DeepSeek and other Chinese AI companies, as well as chip advances by Huawei. Former Republican House Speaker Kevin McCarthy, who chairs an institute that promotes innovation, also carried the message. In early April, when the White House was preparing to cut off H20 sales, Huang attended a $1 million-a-head dinner with Trump at Mar-a-Lago. He told the president that selling the chip in China wasn't a threat to national security. Huang also dangled an incentive: a promise to invest as much as half a trillion dollars worth of AI infrastructure and supercomputers in the U.S. Huang thought he had made a persuasive case, and Trump announced the $500 billion investment. After the dinner, a powerful rival intervened with a counterargument. Musk, still close to Trump at the time, had his own AI ambitions to consider. He told Trump that any advanced chip in Chinese hands threatened U.S. tech supremacy. That view, combined with opposition from national security-minded officials, persuaded Trump to impose the restriction Huang had tried to stop. Huang had been outmaneuvered and didn't enjoy losing—especially when, as he later said, $15 billion in sales were at stake. In early May, Huang spent about two hours with members of the House Foreign Affairs Committee in a private discussion about the threat of Huawei and Chinese competitors and the need for Nvidia to have access to Chinese talent. Later that month, the Trump administration said it was getting rid of the Biden rule limiting chip sales to many friendly countries after pushback from Huang and other industry executives. Huang found a powerful ally at the White House—the so-called AI czar, David Sacks. He shared the fear of ceding the advanced chip market to Chinese rivals. The two men drew the support of Commerce Secretary Howard Lutnick, who oversees export controls and is well-acquainted with tech companies from his former career as a Wall Street dealmaker. All three championed a deal to send Nvidia chips to the United Arab Emirates in mid-May. They later went to see Trump for an unannounced meeting at the White House on July 10. Huang told the president that tight restrictions on Nvidia would just give Huawei a leg up in the AI race. Nvidia needed to access the Chinese market and its AI talent, Huang said: Did the president really want to let China get all the spoils and turn American champions like Nvidia into losers? Commerce Secretary Howard Lutnick, left, and Nvidia CEO Jensen Huang during a White House meeting in April. Trump made his decision, which was announced days later when Huang was in Beijing. China could have the H20 chips as part of a broader trade detente. China hawks who counted on Trump's support were appalled. Democrats on the Senate Intelligence Committee, led by Sen. Mark Warner of Virginia, wrote to Lutnick expressing 'grave concern with the administration's abrupt and inexplicable decision." They said China's commercial AI advances, helped by Nvidia, were 'quickly translating into a national security advantage for the People's Liberation Army." Nvidia has played down concerns about large numbers of chips being diverted to China's military or other adversaries. Beijing would consider such a move risky because Washington could cut off access at any time, according to Huang. 'They simply can't rely on it," He said in a TV interview in July. Winning in Washington was only half the battle. Nvidia's market share in China fell to 50% from 95% over the past four years under U.S. restrictions, Huang said in May. He visited China at least three times this year to reassure Chinese tech executives and government officials that Nvidia was committed to the market. The trips came when many U.S. executives shunned visits to Beijing or even being seen with Chinese tech titans. Huang has met with top executives of Chinese cloud-computing leader Alibaba, smartphone and automaker Xiaomi and OpenAI challenger MiniMax. People in China's tech industry said they appreciated Huang's efforts to modify his chips so they could be sold in China. Engineers there nicknamed him 'Magic Tailor" for his skill in designing chips to thread the needle of U.S. regulations. Jensen Huang surrounded by a crowd in Beijing last month. Knowing the importance of the Chinese market to Nvidia, Beijing increased pressure on the company: China's cybersecurity regulator recently summoned Nvidia representatives to discuss alleged security risks of the H20 chips, citing comments by U.S. lawmakers about the need for a bill to require tracking capabilities for advanced chips sold abroad. Huang and others in the industry have lobbied against the bill, which could face more opposition if it gains traction. The White House has said it wants to study chip tracking but hasn't weighed in on the legislation specifically. Nvidia denied the H20 chip presented such security risks to Beijing. While the H20 can't be used for creating large AI models, such as the ones powering ChatGPT and other chatbots, engineers say the chip is still good at 'inference," the ability of AI programs to tap their training to answer user questions. Since early this year, the H20 chip has played a critical role in filling sky-high demand for AI applications using open-source models such as Chinese-developed DeepSeek and Alibaba's Qwen—especially since Chinese chip makers have had trouble stepping up production capacity. Chinese engineers say companies often get hooked on Nvidia's software, making them reluctant to switch to other companies. That is setting the stage for a battle over Nvidia's next-generation chip that Huang is crafting for China. The company frames the new chip as another stripped-down variant, but it is based on the Blackwell architecture, which is Nvidia's most advanced on the market. Nvidia has begun sending samples to Chinese clients for testing. 'I hope to get more advanced chips into China than H20," the CEO said during his visit to Beijing. 'Whatever we're allowed to sell in China will continue to get better and better." On Monday, Trump said he was comfortable allowing H20 exports because the U.S. has much better chips. The most powerful Nvidia products wouldn't be approved for export unless they were modified for lower performance, he said. Regarding the potential for Huang seeking to export versions of Nvidia's new chips to China, Trump said, 'I think he's coming to see me again about that." Write to Lingling Wei at Raffaele Huang at and Amrith Ramkumar at


Economic Times
40 minutes ago
- Economic Times
OpenAI's GPT-5 release raises revenue risk for Indian IT firms
Getty Images OpenAI may have just fired a warning shot at the Indian IT services model with the launch of the latest version of ChatGPT. GPT-5, ChatGPT's strongest coding version yet released last week, could further dent revenues for an already struggling $283 billion Indian software services industry, as clients may push for even lower pricing. With the upgraded version, the adoption of generative artificial intelligence in software development will rise, exposing IT services firms to revenue deflation risks, Kotak Institutional Equities said in a report. 'We believe GenAI can impact revenue growth for Indian IT by 2-3% for a period of 2-3 years on a net basis,' the report said, adding that the same applies for the customer service and business process outsourcing sector as Indian IT services industry, known for hiring software coders in large numbers, is already seeing the adverse impact of the AI-led disruption. Adoption of the technology has boosted productivity by 20-40%, thereby reducing reliance on engineers and shrinking the people-based billing for Indian IT companies. Besides, the tariffs-induced macro slowdown is also now weighing on the companies that reported negative to flat growth in the last two years.'If your business still depends on armies of coders grinding through routine builds, your margins are about to get hammered,' said Phil Fersht, founder and chief executive of technology consultancy firm HFS Research. 'The smart services players will use this tech to slash delivery times, reinvent pricing and ship services like software products.' Also Read: 'GPT-5 feels dumber': Users on OpenAI's newest model Clients, he said, will push for lower rates as AI boosts productivity, and services providers will need 12–18 months to retool delivery models and replace lost headcount-based revenue with higher-value, AI-augmented services. 'In the short term, expect margin compression before revenue growth,' Fersht launching GPT-5, OpenAI said it is 'our best AI system yet'. According to the company, GPT-5 is a significant leap in intelligence over all its previous models when it comes to coding, math, writing, health, visual perception and more said GPT-5 has substantial advances in reasoning, accuracy and improved instruction following, along with a drop in hallucination rates. This will reduce efforts, costs and risks around legacy modernisation for clients' technology infrastructure providing an edge for IT service providers, they said. Software service providers including large companies like Tata Consultancy Services, Infosys, HCLTech and Wipro are cannibalising their own revenues, as they are hit by a fundamental shift in the pricing model of the industry to outcome-based from employee-based. 'GPT-5 fundamentally lowers the cost structure for enterprise AI by significantly reducing API usage fees,' research firm Gartner said in a separate report.'OpenAI's enhancements — including 65% reduction in hallucination rates, and more transparent uncertainty communications — can reduce compliance risks and make GPT-5 more suitable for enterprise use cases (with the necessary guardrails), it added. Also Read: Is the AI hype justified? OpenAI's GPT-5 will be a potential barometer But not all is doomed. The likes of upgraded GPT versions could accelerate modernisation in the longer-term, aiding IT services firms provide end-to-end services of viable AI-enabled use cases for businesses.'This is the year of advanced models, and GPT-5 is designed to address the needs of developers by reducing hallucinations and simplifying coding through intelligent dynamic routing. While this introduces risks for traditional IT, it also creates new opportunities for innovative AI business use cases,' said Anushree Verma, senior director analyst at also unlocks new enterprise use cases and efficiencies through its improved correctness, steerability and productivity gains. But transformative business value requires thoughtful integration, strong governance and realistic expectations about current limitations, Verma according to Verma, it is important to note that GPT-5 does not achieve artificial general intelligence and is an incremental improvement over previous models. And organisations should remain vigilant and proactive in addressing potential the improvements, its malicious activity support remains similar to GTP-4 and hallucinations continue. 'GPT-5's advanced reasoning, multimodal processing and lower hallucination rates amplify misuse risks, like advanced scam and phishing generation or harmful outputs that risk bypassing restrictions across sessions,' the Gartner report also believes that new opportunities from GenAI adoption will offset revenue deflation in existing volumes over time.'However, we expect a lag in the pickup of new opportunities and for savings in software development to be deployed into these new opportunities, leading to a period of net headwinds,' it said. Also Read: OpenAI offers millions in bonuses to 1,000 employees amid AI talent war Elevate your knowledge and leadership skills at a cost cheaper than your daily tea. As 50% US tariff looms, 6 key steps that can safeguard Indian economy As big fat Indian wedding slims to budget, Manyavar loses lustre Why are mid-cap stocks fizzling out? It's not just about Trump tariffs. The airport lounge war has begun — and DreamFolks is losing Stock Radar: UNO Minda eyeing fresh 52-week high in next few weeks; check target and stop loss for long positions Buy, Sell or Hold: Antique recommends buy on Siemens; Avendus upgrades SBI to Buy post June quarter results Stock picks of the week: 5 stocks with consistent score improvement and upside potential of up to 25% Weekly Top Picks: These stocks scored 10 on 10 on Stock Reports Plus
&w=3840&q=100)

First Post
42 minutes ago
- First Post
Despite Doge cuts, Trump to add over $12 tn to national debt — the highest for any president
Despite coming to office with the promise to cut government expenditure, slash deficits, and reduce national debt, President Donald Trump could add up to $12 trillion to national debt, the highest for any president in American history. read more US President Donald Trump speaks after signing an executive order on creating a White House 2028 Olympics task force in the South Court Auditorium of the White House in Washington, DC, on August 5, 2025. (Photo: Brendan Smialowski/AFP) Donald Trump came to the office with the promise to cut wasteful government expenditure, slash deficits, and reduce the national debt, but he is on his way to add more than $12 trillion to national debt — the highest for any president in American history. The US national debt —the total money that the country owes— has touched $37 trillion and is on its way to surpass $50 trillion over the next decade, with more than a third of upcoming new debt being directly a result of Trump's policies. STORY CONTINUES BELOW THIS AD Such a scenario is in sharp contrast to the promises that Trump made at the beginning of his second term. The Department of Government Efficiency (Doge) fired more than 100,000 government employees, shut entire departments and programmes, and almost killed foreign aid, with the stated objectives to cut government spending and national borrowing. But the overall effect of his policies has been entirely in the other direction. The Congressional Budget Office (CBO) has estimated that Trump's policies, including the 'One Big Beautiful Bill', could add up to $5 trillion to national debt over the next decade. How Trump plunged America in debt like no other president in history Over the next 10 years, Trump's One Big Beautiful Bill (OBBB) will add around $5 trillion to the national debt, according to the CBO's analysis. This includes $4.1 trillion from the OBB, around $3.4 trillion in primary debt and around $700 billion in additional costs, and around additional $900 billion if temporary provisions in the OBBB are made permanent. By 2034, Trump-induced $5 trillion national debt and pre-existing projected national debt growth of $12.6 trillion will take the overall national debt to $54.6 trillion, according to the CBO. As Trump added around $7.8 trillion to the national debt in his first term (2017-21), the total debt in his presidency becomes around $12.8 trillion. This is the highest for any American president. By 2034, Trump's share of overall national debt would be $12.8 trillion out of $54.6 trillion — a whopping 23.44 per cent. STORY CONTINUES BELOW THIS AD