Gridlock Won & Markets Cheer
Ed Yardeni sums up yesterday’s sentiment pretty well:
Financial markets put a positive spin on the blue wave.
Oil prices fell on hopes that the US and Iran are more likely now to resolve their war diplomatically. If so, then the outlook for inflation will improve. So the Fed might not have to raise the federal funds rate as much.
The federal deficit might not widen further if Democrats refuse to increase defense spending, while President Donald Trump vetoes their attempts to provide Medicare for all.
If Democrats slow the AI buildout, hyperscalers’ profits and free cash flow will increase. Fewer AI-related bond issues would also help keep a lid on bond yields. (…)
Investors may be warming up to the idea that some constraints on the AI buildout could be healthy, as we have been arguing recently. Democratic victories in key data center states, including Texas and Georgia, raise the prospect of tighter construction restrictions amid concerns about electricity and water consumption.
Hyperscalers rallied today, with Amazon and Microsoft gaining more than 6%, as capital discipline could ease concerns about AI investment returns and improve free cash flow.
AI infrastructure providers struggled, however. The PHLX Semiconductor Index fell 3.1% amid fears of weaker demand from a slower buildout. We may be entering an AI Goldilocks zone, with enough restraint to take some air out of the bubble without derailing the boom.
This also helped:
“What lifts the green bar is that nothing passes, and a Democratic Congress facing a president with a veto delivers exactly that.” (@AndreasSteno)
- The Election Day Revolt That Threatens America’s AI Boom Americans’ vote for anti-data center candidates and ballot measures could rein in AI’s expansion and hobble economic growth
(…) By Wednesday morning, senior leaders at OpenAI were discussing whether the results meant the company would have to reset expectations regarding its data center capacity next year, according to a person familiar with the conversation.
Americans’ vote for candidates and ballot measures to ban or slow data center development could rein in AI’s rapid growth and become a big business problem for the companies on the edge of technological progress. It also threatens to hobble an AI boom powering the U.S. economy. (…)
Texas, Georgia and Ohio account for more than 40% of all planned data center capacity in the U.S., according to investment bank Evercore ISI. (…)
“This is going to probably elongate the timelines for the data-center build-outs,” Sequoia Capital partner Pat Grady said at the WSJ Tech conference Wednesday. (…)
AI investment is projected to account for 1.9% of gross domestic product this year, according to Goldman Sachs, the biggest share of economic growth for the build out of a new industry since the railroad boom in the 19th century.
(…) roughly 25% to 35% of next year’s data-center build-out in the U.S. could be at risk. (…)
About 56% of voters nationally said data centers were a crucial or major factor in their picks for Congress, according to a national survey conducted by polling firm SSRS. (…)
A decline in new data centers would likely slow down overall construction employment, at a time when building other types of commercial real estate, such as apartments and office buildings, is contracting. (…)
(…) if Wednesday’s broad rally is any indication, Wall Street is betting on a more-predictable, more-“ real” economy in the final years of Trump’s term. (…)
Banks, home builders, carmakers and major retailers ranging from Dollar Tree to Walmart all gained.
“The AI trade was taking up all the oxygen in the economy,” said Michael Antonelli, a market strategist for Baird Private Wealth Management. A slowdown in the data center construction boom “allows the economy to allocate money to other sectors.” (…)
After Vance said on social media that “the reason we lost is the war,” Brent crude-oil futures plunged by 8.7%, to $94.62 a barrel, an apparent sign that traders are dialing back bets on renewed fighting. Government bond yields ticked lower.
Defense contractors RTX, Lockheed Martin and Northrop Grumman also fell. A divided government could make it harder to approve military spending. (…)
We can wonder what the economy/inflation/equity markets would be like without AI and the wars but the reality is that:
- defense spending across the world will not suddenly stop. The US umbrella is gone.
- the AI race will not stop either, not only between OpenAI and Anthropic and others but crucially between the US and China.
- the corporate races to best AI applications is also totally on. This is about competitiveness on products and costs.
- the US elections were not a referendum on AI but on unbridled data center buildout.
Goldman Sachs:
On AI and data center regulation, Democratic wins at the state and local level are likely to have significant implications. Republicans lost the gubernatorial elections in Texas, Ohio, Nevada, Georgia, and Michigan, and Democrats gained a trifecta in Pennsylvania and likely in Arizona as well.
In each of these states except Ohio, the Democratic side explicitly campaigned on stricter data center regulation, stricter data center regulation, and we expect Democratic governors in Texas, Georgia, and Michigan to impose conditional moratoria pending new legislation restricting data centers.
That legislation could end certain tax breaks (a priority for Hinojosa), tighten environmental, water, and electricity rules, and make it easier for localities to impose their own moratoria.
The states where Democrats are likely to further restrict data centers account for roughly 45% of planned US data center capacity, with Texas alone accounting for roughly 25%.
However, since data centers are low on the list of priorities for voters this election, we expect that the issue will fade somewhat in importance in the ensuing months and that these moratoria will not be long-lived if they are enacted.
Demand for compute will keep rising fast and so will competition for data centers.
The elections will accelerate changes already underway towards more rules and a reasonable, pragmatic bargain between data center builders and communities.
In Texas, James Talarico won the the Senate. His victory email made no mention of AI or data centers. But the bigger AI story in Texas was Gina Hinojosa beating Governor Greg Abbott. Hinojosa ran on a moratorium until rules are written, an end to a “corruption tax of subsidies, and protecting local power and water”.
After her victory, Hinojosa said “I don’t hate data centers. I just hate the way they’re taking advantage of Texans,”, leaving good political space for new rules.
The data center industry should rapidly set acceptable standards and/or find new locations.
A significant slowdown in compute supply growth would be disastrous for the US, the US economy and US financial markets.
Morgan Stanley wrote “Democratic victories in state elections, particularly in Texas, could lead to tighter data center restrictions and project delays, although the broader AI investment cycle is expected to stay the same.” Unfortunately, you can’t have “the same AI investment cycle” with data center project delays.
FYI: ChatGPT users went from 1.0 billion to 1.2 billion weekly users with more than 250 million people using its voice features every week. That requires a lot of compute.
From a Deloitte survey released October 29:
- AI is delivering on its promise of speed, with 81% of companies surveyed reporting faster decision-making and execution. 25% say they are using AI to deeply transform the business by creating new products or services, reinventing core processes, or fundamentally changing their business models.
- AI is starting to act, not just advise. Forty percent of enterprises surveyed now allow AI systems to execute decisions or take actions without human approval, at least to a moderate degree, while nearly 1 in 4 state that accountability is unclear when AI produces a negative outcome, making AI safety and governance a growing priority.
- Return on investment (ROI) is expanding beyond productivity, with reported revenue gains from AI increasing to 28% (up from 20% reported in Deloitte’s January report), demonstrating that financial value is solidifying for enterprises. Those companies are also seeing broader benefits beyond productivity and cost savings, suggesting payoff comes not just from moving faster, but working smarter.
More examples that non-AI businesses are slowing:
Qualcomm Slides After Supply Costs Take a Bite Out of Profit
Earnings will be $2.15 to $2.35 a share in the three months ending in December, the company said in a statement Wednesday. The midpoint of that range fell short of the $2.32 estimated by analysts on average, according to data compiled by Bloomberg.
In delivering the forecast, Qualcomm cited a “broad-based increase” in costs for components and manufacturing. Like many chipmakers, the company is dealing with higher expenses caused by shortages in its own supply chain and a dearth of memory chips — largely fueled by the surge in demand for data center equipment. Qualcomm is raising its own prices to cope with the crunch. (…)
Qualcomm’s sales fell 8% to $10.3 billion in the fiscal fourth quarter, which ended Sept. 27. (…) For now, the phone market remains a burden on Qualcomm’s results. Revenue from the segment tumbled 26% last quarter to $5.14 billion. (…)
Arm Holdings Plc, another chip company heavily reliant on the smartphone market, delivered its own mixed results on Wednesday.
Both companies expect the smartphone industry’s shipments to flatten out next year.
Sales at AMD’s data center business nearly doubled to $8.5 billion. Analysts had predicted a total of $8.2 billion. PC and gaming-related sales shrank 8% to $3.7 billion.
US SERVICES PMIs
The headline S&P Global US Services PMI® Business Activity Index remained comfortably above the critical 50.0 no-change mark in October. Although lower than September’s recent high of 58.8 and revised down from the earlier ‘flash’ reading, at 56.1 the index signaled a strong rate of growth that was comfortably above the survey average.
(…) Despite reported headwinds of elevated inflation, high interest rates and associated market volatility, new business inflows rose sharply. Demand was reported to be high, and pipelines of new work were positive with some success converting enquiries into hard contract wins. There was also a net increase in new export business for a third successive month, with growth the best since mid-2023. (…)
Jobs growth was solid, despite being down on September’s 51-month peak and the lowest since July. Companies were keen to expand employment to help support new business and as part of company expansion plans.
However, present capacity was insufficient to keep on top of overall workloads, with backlogs of work rising steeply and to the greatest degree in four-and-a-half years. (…)
An additional challenge facing service providers was another steep increase in input prices. Although easing on September’s near-four-year peak, the rate of inflation was higher than the average for 2026 so far. Companies reported that operating expenses were being driven up by higher fuel prices, which raised prices for a wide range of other goods and services, especially food and associated transportation services. In response to the latest round of increased operating costs, service providers raised their own charges to the greatest degree since July.
Inflation and its impact on client budgets, plus some uncertainty regarding the future path of federal government policies, weighed on service providers’ confidence in October. Although companies expect a strong underlying economy, plus their own plans for greater marketing and company expansion to support growth in the year ahead, sentiment was overall at its lowest since May and was again notably below its typical level.
S&P Global US Composite PMI®
The US Composite PMI remained comfortably above the critical 50.0 no-change mark in October to signal another month of private sector output growth.
That said, amid weaker gains in both manufacturing production and service sector activity, the index fell to 55.9, from 58.4 in September – the lowest in three months.
Private sector new business continued to rise sharply, underpinned by a strong contribution again from services, whilst jobs growth overall remained positive. Cost pressures remained elevated, despite softening since September. Output charges meanwhile rose to the greatest degree since August 2022.
Canada, China Extend Tariff Deal, Aim to Boost Canadian Exports
Canada and China extended a deal to lower Beijing’s agricultural tariffs through the end of 2028, while pledging to continue discussions on boosting Canadian exports to the Asian superpower.
The countries agreed to the extension at a meeting in Beijing co-led by China’s Commerce Minister Wang Wentao, Canadian International Trade Minister Maninder Sidhu said in a news release on Wednesday. (…)
Also, Carney’s office said he’ll travel to China for the APEC meeting in the southern city of Shenzhen that is set for Nov. 18–19. His trip will also take him to Ireland, the Philippines and Cambodia.
The extension means Canadian canola meal, peas, lobsters and crab will face lower Chinese tariffs for another two years. China also restored market access for Canadian beef during Carney’s January visit.
Canada will extend to the end of 2028 certain remission measures — essentially tariff waivers for importers — for Chinese steel and aluminum products not made in Canada. In January, Canada also agreed to import a quota of Chinese electric vehicles at a reduced tariff rate.
“Canada will continue discussions with China to identify opportunities to deliver on the government’s target of increasing Canadian exports by 50% to China,” the government said in the release. (…)