SERVICES PMIs
S&P Global: Business activity growth sustained in January amid a stronger rise in sales
The headline S&P Global US Services PMI® Business Activity Index recorded 52.7 in January, up from 52.5 in December. Remaining above the crucial 50.0 no-change mark, the index has now signaled continuous service sector expansion for exactly three years. That said, growth in January was historically weak and below typical levels seen in 2025.
The rise in business activity was supported by a stronger expansion in new business inflows at the start of 2026. Sales growth was in line with the recent trend amid reports of new client wins and a general improvement in market demand.
The expansion was limited to the domestic market however and even here low consumer confidence and uncertainty tended to limit growth. Regarding international sales, companies recorded a sharp decline in foreign demand, with new export business down to the greatest extent since November 2022 due to tariffs and the uncertain trading environment.
Evidence of ongoing capacity pressures was prevalent during January as outstanding business rose solidly for the eleventh month running, and at the most pronounced rate since last July. Positively, US service providers registered a rise in employment, following December’s slight decline. That said, the rate of job creation was only marginal and weak when compared to the survey’s long-term trend.
Business confidence eased at the start of the year, reaching its lowest since October 2025, reflecting some ongoing political uncertainty in the outlook. Confidence was underpinned by hopes of a stronger domestic economy – in part linked to hopes of lower interest rates – new customer wins and business expansion plans.
Tariffs also remained a key source of cost pressures in January, but overall input prices rose to the slowest degree since last October. Higher supplier costs and payroll expenses also added to upward pressure on company operating expenses.
In line with the trend for overall costs, selling prices increased to a lesser extent than in December, albeit also still higher than the historical average. Firms often sought to pass on their higher input costs to clients, though the rate of growth moderated amid reports of strong competition limiting pricing power.
The S&P Global US Composite PMI® recorded 53.0 in January.
That was up from 52.7 in December and represented a solid rate of growth in private sector activity. Both sectors covered by the survey recorded stronger output expansions, in line with faster gains in new business.
Employment meanwhile rose only marginally, while confidence in the outlook softened. Cost pressures remained elevated, although inflation weakened from the end of 2025. A similar trend was seen for output charges.
Chris Williamson, Chief Business Economist at S&P Global Market Intelligence
“Sustained service sector growth, supported by a robust rise in manufacturing output in January, indicates the economy is growing at an annualized rate of around 1.7%. However, that’s a lower gear compared to the pace of expansion seen prior to December’s slowdown, and hints at GDP growth cooling in the first quarter.
“Consumer-facing companies are increasingly reporting a challenging environment, with demand for services falling in January having nearly stalled in December, reflecting low levels of consumer sentiment and cost of living pressures.
“While financial and business service providers are reporting a more resilient picture, demand growth here is also showing signs of fraying amid heightened concerns over the economic outlook, in turn often blamed on political uncertainty. However, lower interest rates and favorable financial conditions, higher government spending, combined with more active sales and marketing efforts, are propping up business sentiment and spending, and also encouraging modest hiring.
“Inflationary pressures in the service sector meanwhile remain elevated, blamed on the pass though of tariff related price increases and wage growth, though stiff competition is often reported to have limited the impact on final selling prices.”
The ISM: Service Sector Dilemma: Expansion Amid Soft Labor Market & Rising Prices
The ISM Services Index came in at 53.8 in January, unchanged at a 14‑month high first reached in December. The headline strength last month overstated the pickup in demand: new orders shot up, but the rise mainly reflected an unusual dynamic in which fewer firms were reporting declines rather than more firms reporting improvement. That normalized a bit in January. New orders slipped 3.4 points to 53.1, still consistent with an expansion in demand but more in line with underlying conditions than the pop last month suggested.
The select respondent comments struck a slight-tone of optimism for the year. Uncertainty around tariffs continues to cloud the outlook, but a few respondents mentioned stabilizing or potentially improving business activity. AI and data center construction were also key topics highlighted in select comments.
Service sector pricing is an influential factor in the overall rate of inflation. While there is no universal rule about the prices paid component and CPI, a glance at the nearby chart shows that readings about 60 for prices tend to be associated with a rate of CPI inflation that is above the FOMC’s 2.0% target. On that basis, today’s devilish reading of 66.6 is not helpful to the Fed’s efforts keep inflation in check, especially in the context of a dip in the employment component to a near stall-speed of 50.3.
Are Trump’s Tariffs Winning? He says yes, but let’s look at the evidence
(…) Mr. Trump says foreigners bear the costs of the import taxes. He claimed in his essay for us [WSJ] that researchers at Harvard had found that “foreign producers and middlemen, including large corporations that are not from the U.S.” pay “at least 80% of tariff costs.”
We published that claim because readers should know that’s what the President believes, but the paper he cites says something different. In an updated version released after Mr. Trump wrote, the authors note that the “retail pass-through” of the tariffs has been 24%—a measure of the extent to which a given tariff rate feeds through to consumer prices, given that the cost of the good at the border is only one part of the final price. This pass-through rate is higher than under Mr. Trump’s 2018-19 China tariffs.
But that doesn’t tell the full picture of how the tariff cost is distributed. The Harvard economists note in the same paragraph that U.S. consumers are bearing up to 43% of the tariff burden, with U.S. companies absorbing most of the rest. That aligns with other research, such as a recent paper from Germany’s Kiel Institute that found Americans pay 96% of the cost of tariffs. Foreign exporters either pass on the full cost of the tariffs to their U.S. customers, or they ship smaller quantities of goods.
Americans pay one way or the other—via higher prices or less choice.(…)
Mr. Trump also ignores that the tariffs he’s imposing are a far cry from what he proposed. The rates he declared on “Liberation Day” created a market swoon that quickly caused him to back down and promise to negotiate 90 trade deals in 90 days. Some of those have been announced, and most of those are far below his “liberation” rates.
Mr. Trump also quickly made a major carve-out for consumer electronics, including Apple’s iPhones. His full or partial exceptions have since included bananas, coffee, cocoa, jet engines and rare-earth minerals. The President who promised to “drain the swamp” has flooded a new bog with lobbyists seeking tariff exceptions from the trade rep and Commerce Department. And be sure to bring your campaign checkbook.
Perhaps this explains why global retaliation has been relatively mild, and thank goodness for that. Most trading partners have understood that if they sit tight, Mr. Trump might think better of some tariffs.
The big exception is instructive: China. Beijing called Mr. Trump’s bluff with hefty retaliatory tariffs of up to 140% on American goods and a squeeze on rare-earth exports. The result has been a crisis for American soybean farmers who lost their Chinese market—cue hefty subsidies from Treasury—and a scramble by Mr. Trump to sue for trade peace without any behavior change from China.
Harder to quantify is the diplomatic cost of tariffs, as allies conclude the U.S. is an unreliable economic partner. Countries are rushing to negotiate new trade deals with each other without the U.S. China and Southeast Asian countries in October upgraded their previous trade agreement. The European Union and India have signed a free-trade agreement far more extensive than Mr. Trump’s hurried “truce” with New Delhi. The Europe is trying to conclude a major agreement with Mercosur countries in South America. Over time U.S. companies will be at a disadvantage as they compete for foreign markets.
All of this for what benefit? Mr. Trump points to the rising stock market, which is true—but it tends to rise when Mr. Trump dials back a tariff threat, and fall when he issues a new one. The S&P 500 index nearly tumbled into a bear market in the days after Mr. Trump’s April “Liberation Day” announcement of across-the-board tariffs. Stocks saw some of their biggest gains of the year on the days when he announced a pause on the China tariffs, and then a “deal” with Beijing. Tariffs are a market loser.
Mr. Trump also says the tariffs are spurring investment into the U.S. (as a result of his “deals”) and creating a manufacturing renaissance. Back in reality, manufacturing employment declined by some 63,000 jobs in 2025 despite a modest boost early in the year, thanks to reductions in every month after Liberation Day. Some companies have little choice other than to build plants in the U.S. to avoid tariffs, but look for much of that work to be automated. Small U.S. manufacturers in particular have been hurt by higher costs for foreign components that make them less competitive.
Mr. Trump boasts about higher American steel output relative to Japan as a result of tariffs. But employment in steel production has barely budged during his year in office, and employment in industries that use steel such as auto manufacturing is declining.
Higher tariff costs imperil the investments of which Mr. Trump is so proud. Volkswagen’s chief executive recently warned that his company may ditch plans for a new Audi plant in the U.S. You can’t blame him, when his supply chain would be vulnerable to willy-nilly tariffs.
Voters elected Mr. Trump to revive economic growth and tame inflation. His biggest successes have come despite his tariffs, not because of them. He isn’t going to repeal them. But if he froze them in place now and declared victory, he’d have a better chance of persuading Americans that he’s fulfilling his promise.
Karl Rove, senior adviser and deputy chief of staff for President George W. Bush, in the WSJ:
Republicans Have an Economy Problem
(…) The “economy is booming,” he said. It’s been “the best first year of any president ever maybe.”
All this left the impression that the nation’s economic challenges are solved. He made the same mistake President Biden did with the constant refrain that “Bidenomics is working.” Mr. Trump’s declaring that “under my leadership, economic growth is exploding to numbers unheard of” isn’t just exaggerated. It makes people who are suffering feel unseen and abandoned.
Washington consultant Bruce Mehlman described Mr. Trump’s challenge with charts sent to his clients on Sunday. He noted that the University of Michigan’s Index of Consumer Sentiment shows the confidence of Americans without a college degree is the lowest since the question was first asked in 1976. Blue-collar jobs declined by more than 145,000 last year. Low-income households are hit harder by inflation because prices for necessities have grown rapidly. A Fox poll shows Republicans leading among white voters without a college degree by only 10 points.
Add to Mr. Mehlman’s observations that 58% in a new CNN poll say the president’s year was a failure while only 33% believe Mr. Trump cares about people like them. Approval in the RealClearPolitics average of Mr. Trump’s handling of the economy is 41% and inflation 37%.
The president should stop bragging. Many Americans, especially swing voters, feel things aren’t good. A reliable Politics 101 strategy is to explain, empathize, underpromise and overdeliver. (…)
For GOP success this fall, Republicans need a better economic message than what Americans heard from the president in Iowa.
(Bruce Mehlman)
The American and Chinese Economies Are Hurtling Toward a Messy Divorce
(…) Both view their economic competition as a matter of national security.
China’s leaders have determined that disentangling the two economies—often called “decoupling” or “derisking”—is inevitable. The shift fulfills a longstanding Chinese ambition to no longer be a junior partner to the West. (…)
Neither side wants to end all trade between the two economies. But fierce rivalry with the U.S. is now the primary driver of China’s economic strategy, and Xi Jinping is determined to come out on top.
“Over the past year, China has started to see the U.S. as a peer equal,” said Sarah Beran, a veteran American diplomat who is now a partner at Macro Advisory Partners. “China has accepted decoupling, and is now focused on controlling the pace of that decoupling.”
Since early 2024, Beijing has allocated nearly $1 trillion to build self-sufficiency in agriculture, energy and the semiconductors that power its artificial-intelligence drive, a Wall Street Journal analysis of Chinese public records shows. The playbook has already helped China evolve into a powerhouse in sectors like green energy and electric vehicles. (…)
The Trump administration’s 2025 National Security Strategy, a guiding policy document released in December, said the U.S. will “restore American economic independence” and that trade with China “should be balanced and focused on non-sensitive factors.” (…)
China’s share of U.S. imports dropped to roughly 7.5% by late 2025 [from 22% in 2018], according to Goldman Sachs, erasing over two decades of growth following China’s 2001 entry into the World Trade Organization.
China, to compensate, has flooded the rest of the world with cheap goods. It is also routing components to be assembled into products bound for the U.S. through other countries to get around tariffs. China’s annual trade surplus rose to a record $1.2 trillion last year.
Overall, trade between the U.S. and China has plummeted to 2010 levels, according to Moody’s Analytics chief economist Mark Zandi, and investment and tourism are also down sharply in both directions. The two superpowers, Zandi said, are now “running away from each other as much as possible.” (…)
About 9% of Ohio manufacturers in a recent survey said they had reshored some production to the U.S. in 2025, up from 4% in 2021. About 60% of the reshoring in 2025 relocated from China, according to the Manufacturing Advocacy and Growth Network, a nonprofit that conducted the survey.
Ramirez, the Husco CEO, said the company has had some success reshoring production of an electric coil made of copper wire coated with plastic. It found a factory in Michigan and taught workers how to make the part, Ramirez said.
Husco, which supplies Caterpillar, John Deere and big automakers, is resigned to paying the tariffs for other components its U.S. factories need because reshoring them isn’t really possible, Ramirez said. Cast-metal components that require a lot of labor in uncomfortably hot and dusty conditions are one example.
Money and geopolitics are forcing him to keep trying.
“There’s economic pressure to avoid the uncertainty of tariffs in the future by sourcing as much as possible in the U.S.,” he said. Husco owns a large factory in China that makes products for the Chinese market and some non-U.S. destinations, he added.
Tracie Roberts, chief executive of Montville Plastics & Rubber in Parkman, Ohio, said automation and AI are helping her company better compete with China’s lower labor costs. The tariffs on Chinese imports have given Montville another boost, Roberts said, helping it win new business from companies that hire Montville to make plastic items ultimately sold in big-box stores. Montville’s business from such customers has increased roughly 20% since the tariffs, she said.
Plastic products can be made competitively in the U.S. without much problem, Roberts said. Goods that include electronics or many intricate components are more challenging to reshore. And many U.S. manufacturers need help with the cost of adding robots and other automated equipment, she added. (…)
Beyond tariffs, administration officials said the U.S. is preparing a domestic counteroffensive that, through deregulation and new government equity stakes in sectors like semiconductors and critical minerals, will help America eventually gain independence from China in important sectors. (…)
Beijing now calls for “decisive breakthroughs” in six key sectors in the next five years: semiconductors, software, high-end machines, medical equipment, advanced materials and biomanufacturing.
Already, the trillion-dollar push toward self-sufficiency shows that Beijing is effectively attempting to out-invest the West to shore up vulnerabilities.
Spending on semiconductors has ballooned, with $47.5 billion raised in 2024, largely channeled through the China Integrated Circuit Industry Investment Fund, dubbed the Big Fund.
Earlier phases of the Big Fund focused on building chip factories. Now it is transitioning to bankrolling the specialized equipment that remains a foreign chokepoint. In late 2024, the fund channeled roughly $63 million into Piotech Jianke, a subsidiary of the Shenyang-based toolmaker Piotech.
Piotech is experimenting with a workaround for a Chinese weakness in chipmaking. Because the U.S. prevents China from acquiring the most advanced lithography machines needed to shrink chips horizontally, Beijing is betting on Piotech’s “vertical stacking” method, which allows different types of chips—like memory and processors—to be layered on top of each other to increase power and efficiency without needing the smallest, most restricted transistors.
China is spending more on clean energy than any other country, with total investment reaching an estimated $940 billion in 2024, according to Carbon Brief, a British organization that tracks energy and climate issues.
To reduce dependence on imported energy from the U.S. and others, China is planning dozens of nuclear reactors along the coast. Vast hydropower and solar projects are under way in the interior. The urgency has only intensified after the U.S. in January captured Venezuelan leader Nicolás Maduro and threatened a 25% tariff on any country conducting business with Iran—twin shocks that could jeopardize over 1.8 million barrels of China’s daily oil imports.
Perhaps the ultimate geopolitical chip in modern trade is soybeans. Although China is self-sufficient in staples like rice and wheat, its massive pork industry remains dependent on foreign soybeans for over 80% of its feed. If trade routes are blocked, the price of pork—the primary protein for 1.4 billion people—skyrockets, risking domestic instability.
Beijing has shifted some of its soybean purchases from the U.S. to Brazil and Argentina in recent years. Now, it’s increasingly incentivizing production at home, even as it maintains a 25 million-ton annual U.S. purchase commitment as a tactical anchor for the current trade truce.
In the northeastern province of Heilongjiang, where corn has long been the most common crop, the government offered soybean producers subsidies of roughly $739 per hectare—nearly 17 times the amount offered for corn. For local growers, this windfall overrides the market logic that would otherwise favor cheaper imports from the U.S. or Brazil.
Researchers at the Chinese Academy of Sciences are also working to accelerate the breeding of a high-yield line of soybean seeds. Their goal is to close the “yield gap”—the efficiency lead historically held by American farms—by developing seeds that maximize oil content and resist local pests.
Across sectors, China is encouraging its companies to invest abroad, specifically in Africa and Southeast Asia, to diversify supply chains and bypass U.S. tariffs.
The idea, said people close to Beijing, is that separation from the U.S. is acceptable as long as China remains firmly connected to the rest of the world.
Related: Who got a plan?
The “Warsh Cycle”
Trump says the Federal Reserve is ‘in theory’ an ‘independent body’ The president told NBC News in an exclusive interview that he wouldn’t have tapped Kevin Warsh to be the next Fed chair if he didn’t support cutting interest rates.
And while he said he believes the central bank is “in theory” an “independent body,” Trump suggested the Fed should follow his lead because he thinks he knows the economy “better than almost anybody.”
Speaking with “NBC Nightly News” anchor Tom Llamas in an extensive interview, Trump said there was “not much” doubt in his mind that interest rates would soon be lowered. (…)
“We’re way high, OK?” Trump said. “We’re way high in interest. We have now, with me and with all the money, I’m — I’ve always been good at money — and with all the money coming into our country, we’re a rich country again. We have debt, but we also have growth, and the growth will soon make the debt look very small. (…)
Llamas asked Trump whether Kevin Warsh, a former member of the Federal Reserve’s board of governors whom Trump tapped to be the central bank’s next chair, understands that he wants him to lower interest rates.
“I think he does, but I think he wants to anyway,” Trump said. “I mean, if he came in and said, ‘I want to raise them …’
“If he said that, he wouldn’t have gotten the job?” Llamas interjected.
“He would not have gotten the job,” Trump responded. “No.”
Pressed about whether he believes the chairman of the Federal Reserve answers to the president or heads an independent body, Trump said: “Well, I mean, in theory it’s an independent body.”
“But I think, you know, I’m a smart guy,” he continued. “I know the economy better than almost everybody.” (…)
“Now, there’ll be times if I see inflation on the horizon. But we don’t have that,” Trump said. “We have low inf— remember, 1.2% inflation the last three months. We have low inflation, and we have tremendous growth. You haven’t had these numbers like, right? We have — think of it, low inflation, great growth. Now all I have to do is one thing, easy, get down the interest rates.”
The annual rate of inflation in December was 2.7%, according to the Bureau of Labor Statistics’ Consumer Price Index. Throughout the first year of Trump’s second term, the rate of inflation has ranged from 2.3% to 3%. (…)
- Truflation’s US headline inflation today holding below 1% since February, according to real price data: US CPI: 0.93% Y/Y US CPI core (excl energy and food): 1.29% Goods inflation: 0.72% Services inflation: 1.05% (@trueflation)
OpenAI Watch
Google Backs Up $185 Billion in AI Spending With Receipts
(…) “To meet customer demand and capitalize on the growing opportunities we have ahead of us, our 2026 CapEx investments are anticipated to be in the range of $175 to $185 billion.”
$185 billion! Analysts had been expecting something more like $119.5 billion, according to Bloomberg consensus data, which in itself would have been about a 30% increase from 2025. Now investors are being asked to swallow a potential spending increase of more than double. (…)
But what really helped soften the blow of the capex increase were the metrics that pointed to Google’s increasingly formidable position in AI.
As the company moves to integrate its Gemini chatbot into more products like Gmail and the Chrome browser, it is rapidly gaining popularity and eating away at ChatGPT’s market lead. It added 100 million monthly active users quarter-on-quarter, with Chief Executive Officer Sundar Pichai saying the company had also seen a “sharp increase in engagement per user.”
The company has turned what look like a mortal threat — people using AI instead of Google search — into an “expansionary moment” for the core product, Pichai said, with its search business seeing “more usage than ever before.” Google Search revenue was up 17% year-on-year. The company’s recent agreement to provide the technical backbone for Apple Inc.’s redesigned Siri seems set to add to those gains in the not too distant future, Bloomberg Intelligence analysts remarked.
Google Cloud, through which Google sells its AI capabilities and hardware to other companies, grew 48% year-on-year. That extended its run as the fastest growing of the three main providers, having been viewed for a long time as a third player behind Amazon.com Inc. and Microsoft.
This success has been aided by its yearlong investment in its own AI chips, reducing its reliance on expensive and scarce hardware from Nvidia. Its cloud backlog — future agreements for computing power — grew 55% year on year, an increase driven by “multiple customers,” the company said. (Last week, Microsoft said 45% of its backlog was from just one client, OpenAI.)
“Capex spend this year is an eye towards the future,” Pichai told investors on Wednesday evening. “I expect the demand we are seeing across the board, for Google DeepMind as well as for cloud, I think is exceptionally strong.” (…)
- Where is AI showing up in the productivity data? Signs of its effects are already starting to peek through
(…) If AI were helping companies squeeze more out of their employees, you would expect the industries adopting AI most enthusiastically to be enjoying the strongest labour productivity growth.
In the US that correlation has started to show up in recent data. Though of course, correlation isn’t causation and it could be that more innovative industries were most likely to adopt AI in the first place. In a post published by the Federal Reserve Bank of St Louis, some economists try to improve on this analysis in two ways.
First, rather than blunt AI adoption metrics, they ask people to estimate how much time AI tools saved them at work. Second, they look at recent labour productivity growth between the introduction of ChatGPT and the second quarter of 2025, relative to its trend between 2015 and 2019.
This was supposed to strip out any pre-existing trends that could mess up the results. Combining these two metrics, they found that the industries where workers were saving the most time using AI were also the ones seeing unusually fast labour productivity growth. These included information services as well as professional, scientific and technical services.
And updating the data to the third quarter of 2025, it looks like the correlation strengthened slightly. I wouldn’t take the self-reported time savings too literally, not least because not everyone is as diligent as me, reallocating the time I save using ChatGPT (to find data) towards making my output even more jolly. (…)
It’s also reasonable to be sceptical of these correlations because LLMs have only recently graduated from “precocious 11-year-old” to “cocky graduate intern”, and towards the end of 2025 self-reported AI adoption by US businesses was still below 20 per cent.
So my final bright spot comes in the form of a study taking a longer view of the data and a broader view of the technology. Jonathan Haskel, one of the study’s authors, explained that 2017 was the real technological turning point, when a famous “deep learning” paper introduced the transformer architecture in machine learning (the “T” in ChatGPT), boosting generative AI. Which is why they compare the period between 2017 and 2024 with the one between 2012 and 2017.
More specifically, the authors study US investment in software and estimate how much it has contributed to growth. This involves various assumptions, as they try to include both the productivity gains associated with companies becoming better at producing the software, as well as the effects of other industries using it. They estimate that together, these contributed as much as half of the increase in productivity growth between the two time periods.
All of this is suggestive — the sun clearly hasn’t come out fully yet. We don’t have the data to repeat that last analysis in Europe. And when senior McKinsey adviser Tera Allas examined the British data, she couldn’t find any evidence that AI-adopting industries were experiencing unusually high productivity growth. Still, I’m trying to stay positive. Otherwise maybe you will decide that your columnists are better in artificial form.
China Ramps Up Energy Boom Flagged by Musk as Key to AI Race
New data on China’s relentless energy installations underscore warnings from Elon Musk and Jensen Huang that the nation’s world-beating power network will deliver a major advantage over the US in the race to dominate artificial intelligence.
Since 2021, China has added more power capacity across all energy technologies than the US has in its history, including 543 gigawatts last year, according to figures released late last month by the country’s National Energy Administration.
China will add more than 3.4 terawatts of electricity generation capacity over the next five years, almost six times as much as the US, BloombergNEF projects under its base-case Economic Transition Scenario. That influx would give the No. 2 economy greater ability to accommodate rising power demand from data centers.
“The limiting factor for AI deployment is fundamentally electrical power,” Musk told BlackRock Inc. Chief Executive Officer Larry Fink in an interview at the World Economic Forum on Jan. 22. “Very soon, maybe even later this year we’ll be producing more chips than we can turn on — except for China,” said Tesla Inc. CEO Musk, whose xAI is building US data centers. “China’s growth in electricity is tremendous.”
Musk’s comments echo similar sentiments from Nvidia Corp.’s CEO Huang, who has also identified access to electricity as a potential differentiator between the US and China. AI competitiveness can be thought of as a cake composed of five layers: energy, chips, infrastructure, models and applications, Huang told a December event hosted by the Center for Strategic and International Studies. “At the lowest level — energy — China has twice the amount of energy we have as a nation,” he said.
In the US, data centers will account for 38% of the growth in electricity demand between 2024 and 2030, though just 6% in China, according to BNEF projections. Data centers will command almost 7% of total US power demand by 2030, compared to 2% in China.
China, where electricity consumption hit a record last year, is continuing to rapidly add new power capacity with huge deployments of renewable sources like solar and wind alongside more coal, nuclear and gas facilities. Solar generating capacity is on track to surpass coal in China for the first time this year, though newly added renewables have lower utilization rates than fossil fuel assets.
Growth in generation capacity is proving slower in the US, where electricity demand has recently begun surging after roughly two decades of flat growth through the early 2020s. The anticipated requirements from AI have triggered a rush of development of new gas-fired capacity, though US power plants can take years to come online as a result of more onerous regulation and supply chain bottlenecks. President Donald Trump’s opposition to renewables also means clean energy projects have been subject to cancellation or delays.
“The federal government is essentially shooting ourselves in the foot by not allowing a more straightforward and robust uptake of” renewables to meet data center demand, said Michael Davidson, an associate professor at UC San Diego who focuses on US and Chinese energy policy.
Already, snarls across the US power grid are becoming a drag on the AI industry. Some US utilities are telling data centers it will take years for them to connect, while northern Virginia — a major global hub for the sector — has experienced instances of disconnections<?XML:NAMESPACE PREFIX = “[default] http://www.w3.org/2000/svg” NS = “http://www.w3.org/2000/svg” /> as a result of grid faults. The mismatch of AI power demand and slower growth in capacity additions could trigger effective electricity shortfalls in some US markets by 2030, BNEF said in a December report.
In contrast, connecting to the grid is “a non-issue in China” for new data centers, according to David Fishman, a Shanghai-based principal at consultancy The Lantau Group.
China is likely to have spare power capacity equal to more than three times the world’s entire data center demand by 2030, according to Goldman Sachs Group Inc. analysts. A lack of similar capacity “could act as a bottleneck for further data center developments in the US,” analysts including Hongcen Wei wrote in the November note. (…)
“The AI industry talks about time to power because that’s what’s rate-determining for them,” she said. “It’s not chip availability, it’s availability of power.”
AI will account for a smaller share of total power demand in China than the US in part because of the impact of the Asian nation’s other energy-hungry sectors, including industry and electric vehicles — both of which will add electricity consumption at a faster rate through 2030.
Still, energy access isn’t the only factor that’s likely to determine success in the AI sector, and the US retains its own advantages. China’s AI companies haven’t yet been able to innovate beyond the current cutting edge of technology and remain about six months behind the frontier AI of the most innovative Western firms, according to Google DeepMind CEO Demis Hassabis.
“China has an abundance of energy,” said Gartner Inc. analyst Chirag Dekate. “I think the US has an innovation edge in the chip layer and in the model layer.”
Hmmm…
(…) Nearly 75% of the power equipment planned to be used on site at data centers is natural gas, according to a report released Tuesday by Cleanview, a market intelligence platform. (…)
“This is a very new trend,” Cleanview’s Michael Thomas wrote recently. “A little more than a year ago, virtually all data center developers planned to use the electric grid to power 100% of their projects.”
Wind, solar, and batteries are cheaper in a lot of places. But the AI race is so intense that companies are choosing power they can get now over power that’s cheaper later, amid years-long grid connection delays. (…)
Data: Cleanview data center tracker. Chart: Axios Visuals
