U.S. Adds 115,000 Jobs in April With Solid Hiring Across Sectors

The U.S. job market blew past expectations again in April, buoyed by gains across industries including retail, transportation and warehousing, and healthcare. The results were a sign that the labor market remained resilient so far in the face of the Iran war.

The American economy added 115,000 jobs in April, the Labor Department said Friday, far exceeding expectations.

That was down from a net gain of 185,000 in March. But it was much better than the 55,000 jobs that analysts polled by The Wall Street Journal had expected to see for April.

The unemployment rate stayed unchanged at 4.3%, as economists had expected. (…)

In the first four months of the year, monthly payrolls have averaged 76,000, up from an average of about 42,000 during the same period last year. (…)

The Friday report should put the focus squarely on inflation data when it comes to determining where the Fed—now firmly on pause—goes from here. While the labor market has steadied, inflation is now drifting up instead of down due to the effects of tariffs and the Iran war, which has sent gasoline prices soaring.

With the labor market giving the Fed cover to wait, the next move on the policy debate is when and how to tilt toward a neutral bias that suggests a rate increase could be as likely as a rate cut. The answer could turn almost entirely on the inflation numbers. The Labor Department will report the latest data on the consumer-price index next week. (…)

A key measure of underemployment also rose—a sign more people are taking on part-time roles because they can’t find a full-time job.

The U-6 rate, which includes people working part time who would prefer a full-time job and people who are so discouraged they have stopped looking for work, rose to 8.2% in April from 8% in March. It is at the highest level since December and more than a percentage point above where it was just before the pandemic.

Closely watched by the Fed, U-6 suggests many Americans are struggling with underemployment. The number of people working part time who would have preferred full-time employment increased by 445,000 to 4.9 million in April, the Labor Department said. (…)

The household survey is not quite as hopeful as the establishment survey, the latter being down for the 4th consecutive month during which time it declined by 343k monthly jobs on average to reach its lowest since December 2024.

Smoothing out the monthly establishment data, the 3-m pace of hiring is now 48k vs the 6-m average of 55k.

A slow May, or revisions, could reset the “resiliency” narrative.

The U-6 unemployment rate is still creeping up, as are the number of part-timers, as Americans find ways to make ends meet. At 8.2%, the U-6 is well above its Q4’2029 level of 6.8% while part-timers are up 16.5%.

Hourly earnings rose 3.6% YoY, in line with its 4-m average. But MoM, growth has slowed from +4.3% annualized in January to +2.6% in March and to +1.9% in April.

Indeed Job Postings peaked in early March and are still weak through April 30, suggesting the next JOLTS report will be weak.

Canada: April’s jobs report reinforces the weak start of the year

From LSEG IBES:

440 companies in the S&P 500 Index have reported earnings for Q1 2026. Of these companies, 83.2% reported earnings above analyst expectations and 13.2% reported earnings below analyst expectations. In a typical quarter (since 1994), 67% of companies beat estimates and 20% miss estimates. Over the past four quarters, 78% of companies beat the estimates and 17% missed estimates.

In aggregate, companies are reporting earnings that are 8.1% above estimates, which compares to a long-term (since 1994) average surprise factor of 4.4% and the average surprise factor over the prior four quarters of 7.1%.

Of these companies, 78.2% reported revenue above analyst expectations and 21.8% reported revenue below analyst expectations. In a typical quarter (since 2002), 63% of companies beat estimates and 37% miss estimates. Over the past four quarters, 73% of companies beat the estimates and 27% missed estimates.

In aggregate, companies are reporting revenues that are 2.1% above estimates, which compares to a long-term (since 2002) average surprise factor of 1.3% and the average surprise factor over the prior four quarters of 1.9%.

The estimated earnings growth rate for the S&P 500 for 26Q1 is 28.6%. If the energy sector is excluded, the growth rate improves to 30%.

The estimated revenue growth rate for the S&P 500 for 26Q1 is 11%. If the energy sector is excluded, the growth rate improves to 11.6%.

The estimated earnings growth rate for the S&P 500 for 26Q2 is 22.9%. If the energy sector is excluded, the growth rate declines to 19.6%.

Revisions are exceptionally strong:

 

Amid a highly uncertain world, many more companies are offering guidance than at the same time last quarter (86 vs 59), slightly more negative (+51%) than positive (+33%).

Trailing EPS are now $286.72, up 4.1% from Q1 and +12.9% YoY (S&P 500 Index is up 24.2% YoY). Full year estimates: $336.49e (+24.0%). Forward EPS: $347.01, up 7.7% from Q1 and +28.8% YoY). Full year 2027e: $386.7 (+14.9%).

Since 2019 (pre-pandemic) S&P 500 EPS would be up 106% by the end of 2026. Doubling in 7 years!

That’s only happened in 6 periods since the Great Crash of 1929, the last ones in 2015-17 and 1998-2001.

But hang on here!

US public companies are required to report the quarterly mark-to-market value of their equity holdings in their net income.

In early 2026, the mark-to-market rule has already caused massive, multi-billion dollar swings in the reported net income of several S&P 500 giants, effectively detaching their “headline” earnings from their actual operational performance. E.G.:

• Alphabet (GOOGL): Reported a massive $37.7B gain in other income for Q1 2026, primarily from net unrealized gains on its non-marketable equity securities (including stakes in Anthropic and SpaceX). This non-operating tailwind added roughly $2.35 to its diluted EPS, making its 81% net income jump look much larger than its 30% operating income growth.
• Amazon (AMZN): Its Q1 2026 net income included $16.8 billion in pre-tax gains from its investment in Anthropic.
• Meta recorded $6.9B in “income tax benefits net of investment losses”.

Goldman Sachs notes that while the aggregate S&P 500 earnings growth was reported at roughly 25% so far, “normalizing” GOOG and AMZN results would bring the underlying earnings growth was closer to 16%.

We are not done normalizing. Many other companies yet to report will also include gains in their Anthropic/OpenAI/SpaceX/etc. holdings as of Q1, but also going forward when more gains will likely accrue.

Be careful looking at some P/E ratios! The “E” may not be sustainable operating earnings.

Based on recent filings, GOOG owns 14% of Anthropic, AMZN between 15-21%, MSFT 5-10% and NVDA 3%.

Alphabet and Amazon used Anthropic’s February 2026 Series G funding valuation of $380B for their Q1 mark-to-market.

Last week: Anthropic weighs deal for near $1tn valuation as revenue surges

The new round is expected to value Anthropic at about $900bn pre-money and to raise as much as $50bn, said three of the people. They added it was likely to close within two months. OpenAI was valued at $852bn post-money in March after it closed a record funding round of $122bn.

FYI, based on my available info:

• There are 12.1B GOOG shares outstanding, meaning GOOG’s 14% stake would be worth $125B at $900B valuation or $10 per GOOG share.
• There are 10.7B AMZN shares outstanding. Its ~18% approximate share would be worth $162B at $900B valuation or ~$16 per AMZN share. AMZN also owns ~6% of OpenAI, worth some $51B or ~$5 per AMZN share.
• There are 7.4B MSFT shares outstanding. Its ~7.5% approximate share would be worth $68B at $900B valuation or ~$9 per MSFT share. MSFT also owns ~27% of OpenAI, worth some $230B or ~$31 per MSFT share.

Back to earnings:

Q2 estimated growth is 22.9%, thanks to Energy (+100%), Materials (+30.6%) and Tech (+55.4%).

All 7 other sectors are seen up 8.7% on average vs +9.3% on April 1 and 10.0% on Jan. 1. The term “broadening” market may be too broad.

Mind you, 8-9% profit growth is good, but we need to monitor the slowdown given the rising squeeze inflation is having on consumers’ real spending power.

Weekly payrolls were still up 4.0% YoY in April but PCE inflation was up to 3.5% in March, rising fast from +2.8% in since last November.

Putting aside hours worked (cancelling every second month), payrolls growth has been slowing every month since January, increasingly relying on wage growth (black) which is also slowing fast as seen above.

About the broadening market:

Big Tech’s heft is obscuring emerging weaknesses elsewhere.

According to a recent UBS analysis, 42 stocks are driving the bulk of the S&P 500’s returns; typically, around 100 do.

The index is up 12 per cent since the end of March on the back of AI-fuelled tech blue-chips, which also include Alphabet, Microsoft, Apple, Meta Platforms and Broadcom. The equal-weighted version of the benchmark, which in effect turns down the volume on those giant stocks, is up half as much.  (FT)

 

Topdown Charts Professional

• Never mind valuations, Big Mo is the game:

• The below helps explain the above…

 @MikeZaccardi

• … and some of the below:

Goldman Sachs

• Positioning at extremes!

Topdown Charts Professional

Ed Yardeni explains what’s currently happening:

(…) late last year, a growing concern that the AI boom was turning into an AI capital spending arms race among the Mag-7 was heightened by Michael Burry’s warnings that the hyperscalers’ massive AI capex might prove unprofitable for various reasons.

But those concerns have diminished in response to significant beats by the hyperscalers during the Q1 earnings season in April. Their cloud earnings continue to soar, confirming that rapidly growing demand for “compute” might justify all the AI capital spending after all.

(…) Here is a quick review of how the AI-11 fits into the AI supply chain:

(1) Foundry and lithography (TSMC, ASML). TSMC fabricates the leading-edge logic for everyone. ASML owns the EUV lithography chokepoint.

(2) Logic and custom silicon (AMD, Broadcom, Intel). AMD is taking a significant share in AI inference. Broadcom is the custom ASIC partner for hyperscalers and the incumbent in networking silicon. Marvell rounds out its custom silicon, networking, and optical connectivity offerings. Intel is the foundry comeback story with CPU exposure to the AI server cycle.

(3) Memory (Micron, SK Hynix, Samsung). Micron, SK Hynix, and Samsung supply the high-bandwidth memory that is the actual bottleneck for AI training. SK Hynix leads the HBM market globally.

(4) Enterprise NAND and storage. SanDisk has emerged as the pure-play beneficiary of NAND and enterprise SSDs. Western Digital provides the HDD complement.

Every dollar of hyperscaler capex for AI infrastructure flows through this supply chain before reaching a server rack.

David and I have been strong believers in AI. The February 9, 2026 post Railroaded? and the March 13 deep dive The AI Supercycle: A Deep Dive offered our supporting analysis. 

Concluding Railroaded? I wrote:

• AI is truly transformational and will be quickly widely adopted for productivity and competitiveness imperatives.
• Unlike the railroads and the IT infrastructure booms, there is little front-loading “hoping/waiting for demand” investments, nor much leveraging at this time.
• Prices/costs are coming down fast, necessary to boost demand/usage along the way.
• Agentic AI will be huge.
• In the AI era, moat is crucial and bigger is better.
• Diversified revenue/cashflow streams are desirable.
• No or low indebtedness is preferable.

From a macro perspective,

• AI is clearly and significantly boosting GDP growth.
• AI is clearly inflationary in some sectors (construction, commodities, power), disrupting other sectors by hording resources.
• The massive hyperscalers expenditures (almost the size of the US defense budget) are redirecting their excess cash from the fixed income markets into the real economy with high multiplier effects. Will productivity gains offset demand-pull inflation?
• Will these financial flows (reduced corporate demand) impact US interest rates?

Only 4 months later, demand is even stronger than expected thanks to Anthropic’s Claude models. Compute demand is even too strong, too quickly, forcing Anthropic (and some others) to tame demand through higher prices and controlled access until more data centers add to compute supply.

The various expected roadblocks have materialized, e.g. power (Power Play Sept 23, 2024), equipment, workers and NIMBY protests, slowing supply growth in the short-term but extending the buildout cycle.

Costs are rising fast, magnified by the US war on Iran and its effects on various supply chains. An increasing part of AI capex is scarcity-induced inflation, positive for suppliers’ margins dealing with scrambling buyers, largely price insensitive.

BTW, this is a dangerous situation for inflation. Consider:

• Exploding AI compute is fueling significant demand other than for chips and memory for resources such as power (oil, gas, solar, nuclear), labor (construction, plumbers, electricians) and specialized hardware (turbines, transformers, cooling equipment). Time is of the essence so hyperscalers are totally price insensitive, a situation likely to persist through 2030.
• Hormuz is forcing countries and corporations to build reserves and hoard various critical resources and materials as protection from constrained supply chains. Hoarding of oil, gas, helium, fertilizers, chips, etc. is now required in an increasingly selfish, uncooperative environment. This will permanently raise resource costs across the world and reduce/eliminate operating efficiencies built during globalization.
• The Ukraine and Iran wars have significantly depleted ammo and military equipment inventories which will need to be hastily rebuilt. At the same time, the world has also learned that alliances can be fickle, incentivizing many countries to find ways and means to protect themselves. Demand for military equipment will be firm and largely price insensitive for several years.

AI and productivity will not help offset the huge, world-wide demand for such a broad spectrum of resources and goods against largely price insensitive buyers.

This was already evident from April’s J.P. Morgan Global Manufacturing PMI:

• Global manufacturing growth accelerates in April despite sharp rises in prices and supply chain delays

The start of the second quarter of 2026 saw rates of expansion in global manufacturing output and new orders strengthen. However, price and supply chain pressures continued to build (…).

Manufacturing production increased for the ninth month running, with the rate of growth hitting a near five-year high. Expansions were signalled across the consumer, intermediate and investment goods sectors.

The increase in production was supported by faster growth of new business. (…)

The resulting input shortages and delivery delays led to a solid upswing in purchase price inflation, as demand exceeded available supply. Average input costs rose at the quickest pace since June 2022 and at one of the fastest rates in the 28-year survey history. (…)

Average output charges rose at the sharpest rate in 45 months, as manufacturers passed on part of the increase in costs to clients. Backlogs of work meanwhile rose for the third month running.

As Ed Yardeni illustrates, the recent jump in the ISM-Prices Paid Index almost guarantees higher PPI inflation …

… which would translate into higher consumer inflation …

… potentially scaring central bankers:

Oil Market in ‘Race Against Time’ on Hormuz, Morgan Stanley Says

(…) “The United States’ 3.8 million barrel-a-day increase in exports and China’s 5.5 million barrel-a-day cut in imports have shielded the rest of the world from 9.3 million barrel-a-day of tightness — a very significant amount,” the analysts said in a section headlined “A race against time.” (…)

“The path matters: a reopening in June with US and Chinese buffers still partly intact is the base case; a closure that runs into late June or even July is the regime in which Brent flat price has to do work it has so far been able to avoid,” they said, referring to futures for the global crude benchmark.

In the bank’s still-current base case, Dated Brent — a physical marker — is seen at $110 a barrel this quarter, $100 in the following three months, and $90 between October and December, with forecasts unchanged. In the bull case — based on a longer closure — prices were seen at $130 to $150. (…)

“Even if the strait reopened tomorrow, the time required to restart fields, repair refineries and reposition tanker tonnage means the market is on track to lose another billion barrels over the balance of 2026,” they said.

But for Ed Yardeni, the race is against earnings:

Raising Our 2026 S&P 500 Target Range Due To Earnings-Led Meltup

We are raising our year-end S&P 500 target from 7700 to 8250. We’ve been bullish on earnings but not as bullish as the recent consensus of industry analysts. We’ve never seen consensus earnings expectations rise so quickly for the current and coming years as they have in recent months. The result has been an earnings-led meltup in the stock market.

Our 2026 and 2027 EPS estimates have been $310 and $350, respectively, since late last year. Those were bullish estimates back then. Consensus EPS estimates have rocketed above our targets in recent weeks. They are currently $336.49 (up 22.0% from last year!) and $386.70 (up 14.9% from the 2026 consensus estimate).

We are raising our EPS estimates to $330 this year and $375 next year (Our $375 year-end estimate for forward earnings is conservative, in case the analysts are too exuberant). We are sticking with our forward P/E range of 18.0-22.0, resulting in a year-end range for the S&P 500 of 6750-8250, assuming (as we do) that forward earnings per share will be will be $375 at the end of this year. The latter is already at $354. (…)

Our outlooks for EPS and RPS imply that the S&P 500 forward profit margin will rise to 15.0 this year and 16.3 next year. These forecasts are a bit higher than the current consensus. (…)

We are now also raising our subjective probability of a continuation of the Roaring 2020s to 80% from 60% simply by merging it with our meltup scenario (previously at 20%). We are doing so because we believe that any meltdown will be a buying opportunity and won’t trigger a recession or bear market similar to the 1999-2000 Tech Bubble and Tech Wreck. We are sticking with 20% odds of a recession that causes a bear market. (…)

What could possibly go wrong? The war in the Middle East isn’t over, though that hasn’t kept stock markets from soaring around the world in April and so far in May. That’s because oil prices have remained around $100 per barrel. The shock waves from the war could still hit the global economy. Another round of fighting could be even more troublesome, as it could result in stagflation. A more persistent inflation problem would force central banks to raise interest rates. The Bond Vigilantes would likely push bond yields higher in this scenario.

Nevertheless, for now, we are sticking with our 10,000 target for the S&P 500 by the end of 2029. It might arrive ahead of schedule.

Productivity Booms As Labor Market Shows Signs Of Revival

(…) on balance, the latest batch of labor market data suggests that employment conditions may be improving (…).

We [Ed Yardeni] agree with Jevons’ Paradox: Making a production input more efficient lowers the cost of the final product, stimulates demand for it, and ultimately results in greater demand for the input itself, despite the productivity gain.

Productivity is measured as nonfarm business output divided by labor hours worked. Output increased 3.3% y/y in Q1-2026, solidly above the comparable 2.7% rise in real GDP. Hours worked rose only 0.4% y/y.

So productivity increased 2.9% y/y, exceeding its historical average of 2.1%.

In our Roaring 2020s scenario, productivity growth is likely to increase to 3.5%-4.0% over the remainder of this decade and continue at that pace through the Roaring 1930s.

Unit labor costs is measured as hourly compensation divided by productivity. It rose by 1.2% y/y in Q1-2026, the slowest pace of growth since Q3-2023.

This confirms our view that the labor market isn’t currently a source of inflation but rather disinflation. For now, the latter is being offset by other inflationary pressures, i.e., higher energy prices and tariff-related increases in durable goods prices.

Inflation-adjusted hourly compensation is determined by productivity. Businesses can only sustainably raise real pay when productivity gains provide the underlying economic value. We expect productivity growth to rise close to 4.0% by the end of the decade, supporting equivalent real hourly compensation growth (chart).

Strong productivity growth tends to widen profit margins. Profit margins are currently at record highs, and boosting corporate profits. S&P 500 earnings growth has been surprisingly strong as a result.

The May 7 Challenger, Gray & Christmas report showed that US employers announced 83,387 job cuts in April. For the second consecutive month, AI was cited as the primary reason for layoffs, accounting for 26% of all cuts (roughly 21,490 jobs).

Despite the monthly jump, year-to-date layoffs remain down 50% compared to the same period in 2025. This confirms that while specific sectors are being disrupted by AI, the overall labor market remains resilient. It is also entirely consistent with our belief that AI will create jobs on net.

Yesterday, ADP reported that private-sector payrolls rose by 109,000 in April, the fastest pace of job creation since January 2025. The result is corroborated by Revelio Labs, which reported that total nonfarm payrolls rose by 66,400 in April, the most since March 2025, with gains led by health care and social services and the finance sector.

Yardeni here uses YoY growth rates (left chart) which is helped by a weak Q1’25. On a QoQ basis (right), productivity growth has slowed since Q3’25 to +1.6% annualized in Q4’25 and to +0.7% in Q1’26.

 

Since Q4’19, labor productivity has grown at an annualized rate of 2.1%.

Ed also focused on the YoY Challenger layoffs data, helped by the jump in tariff-induced levels in April of last year.

Challenger does not publish seasonally adjusted numbers but notes that job cuts in April were up 38% from March and that this April’s total is the third highest since 2009.

David Rosenberg’s own seasonal adjustments show layoffs rising from 37k in February to 40k in March and to 74.5k in April.

Also:

Hiring plans fell 69% in April to 10,049 from 32,826 in March. They are down 38% from the 16,191 hiring plans announced in April 2025. So far this year, employers have announced plans to hire 60,936 workers, down 13% from 70,058 new hires announced during the same period in 2025.

“With a number of factors potentially impacting summer travel plans as well as how businesses operate across sectors, we predict hiring plans will remain muted,” said Challenger.

Goldman Sachs’ wage tracker stands at 3.1% annualized in Q1 (vs. 3.7% in Q4) and 3.6% year-over-year (vs. 3.7% in Q4).

PCE inflation was +3.1% YoY in Q1, +3.5% YoY in March. On a QoQ basis: +4.5% annualized in Q1, +8.3% in March.

This seems like a good place to insert this:

Ray Dalio Says US Is Entering Period of ‘Great Turbulence’

Bridgewater Associates founder Ray Dalio said the US is headed for years of tumult, driven by large deficits, the growing wealth gap and left-right political divisions.

A new geopolitical system and disruptions from AI will also contribute to the turmoil, he added.

“There will be huge changes over the next five years, with all of these forces coming together,” Dalio said in an interview on the New York Times podcast Interesting Times with Ross Douthat. “And on the other side of that, it’ll be almost unrecognizable. It’ll be very different, and it’ll be a period of great change and great turbulence.” (…)

“We’re going to come into the midterm elections and I think that the Republicans will probably lose the House,” he said. “I think from that point on, you’re going to see an intensification of political and social conflict that’ll take place in that period, particularly between that election and the presidential election in 2028.”

Internationally, he said, there is no longer a rules-based order and the outcome of the US-Iran war can be defined in “almost black-and-white terms of who will control the Straight of Hormuz, and who will control the nuclear materials.”

Investors should maintain a well-diversified portfolio with between 5% and 15% in gold, he said. “When we look at history, we see that in all such periods, all the fiat currencies go down, and gold goes up.”

Separately, Apollo Global Management Inc. Chief Executive Officer Marc Rowan warned Thursday of a “massive geopolitical realignment” that will lead to “blue-collar ascendancy and white-collar stress.”

Yesterday, David Rosenberg published these 2 charts. I added the black dashed lines to show where the US was before Trump 1.0 vs the pre-1970’s period. Since 2018, the corporate sector has hugely outgrown the personal sector, creating the largest imbalance ever.

 

• BofA’s Hartnett Says US Stock Gains Have Rarely Been This Strong

(…) US stocks and gold are hurtling toward a fourth year of double-digit gains, rallies so exceptional that there have been few such episodes in history, according to Bank of America Corp. strategists.

The S&P 500 is set for an annualized 20% gain while bullion is on track for a 30% rally, said the team led by Michael Hartnett. For stocks, such prolonged, “big stuff” advances only played out in World War II, the period of peace that followed a few years after that conflict and in the bubble of 1995-1999, they said. (…)

The latest impulse from the artificial intelligence capex frenzy has pushed the US market higher in a very narrow rally driven by a just a handful of stocks. Now, other areas have started to see stronger gains. (…)

Consensus forecasts now predict the American economy to expand by a nominal 5.5% this year, with earnings growth at 20%.

The BofA team tipped material stocks to be the next strong gainers. While this sector accounts for just 2% of the S&P 500, close to 30-year lows, this is set to change.

A geopolitical grab for resources, increased military spending, the AI capex boom and efforts to address housing shortages should make materials “the new bull on the block,” they said.

Meanwhile:

BlackRock Cut Its Private Credit Fund Five Percent.

Two days ago, Oaktree (OCSL) marked down its private credit fund and cut its dividend. Today, BlackRock TCP Capital (TCPC) did the same thing. Two major asset managers, same week, same category.

BlackRock cut its fund value 5 percent. Six portfolio companies drove most of the decline. The common thread was software loans written in 2021 when valuations were high and rates were near zero. Those same companies now carry higher debt costs and lower revenue as AI replaces their products.

That double pressure was not in any 2021 underwriting model. It is in the 2025 results.

Gundlach warned at Milken this week that private credit investors will lose money and compared the market to dot-com and pre-crisis mortgages.

Two funds confirmed. The next two weeks of BDC earnings reveal whether this is a pattern or a coincidence. One more major fund markdown makes it a pattern.

The data is now catching up to the warning.

Much more than microchips: Trade soars in AI-related goods, driving U.S. trade deficit

American imports related to the artificial intelligence boom have more than doubled since 2023 while non-AI imports have fallen. This hunger for AI investment has increased the U.S. trade deficit despite the highest tariffs in a century. AI-related exports from the U.S. have also surged, though not as much, according to new (AI-assisted) research from Minneapolis Fed Monetary Advisor Michael Waugh (Minneapolis Fed Staff Report 684, “Trade in AI-Related Products”).

By tapping a large language model (LLM) to analyze data for more than 18,000 products, Waugh pushes beyond obvious AI inputs, like computer hardware, to capture the broad range of goods related to the build-out of data centers and other infrastructure. It’s not just microchips and circuit boards: A sharp rise in categories such as electric power, cooling HVAC, and telecommunications clearly accompanies the recent investment in AI. (…)

As of January 2026, the analysis finds imports of AI-relevant goods were 111 percent higher in nominal dollars than the monthly average in 2023. Imports with low AI relevance were down 14 percent. Prior to 2024, these high and low AI-relevant bins displayed nearly identical trends.

Another way to quantify the change: For all of 2025, total dollar imports of AI-relevant products ($379 billion) were 72.6 percent higher than in 2023. Imports of products with low AI relevance rose just 2.5 percent.

This growth occurred alongside historic increases in U.S. import tariffs during 2025. Using methods from his tariff-related research, Waugh calculates that AI-relevant goods faced an effective tariff rate of just 4.5 percent, versus 12.1 percent for non-AI goods. Tariff exemptions—principally one for consumer electronics—covered about 69 percent of AI-relevant imports, a situation that remains largely unchanged after February’s Supreme Court rule invalidated some tariffs.

AI trade flows both ways. U.S. exports related to AI were 35 percent higher in 2025 than in 2023. But the rise in imports was much larger. In a counterfactual exercise, Waugh finds that without these AI effects on trade, the U.S. trade deficit would have been 16 percent smaller in 2025.

The analysis unsurprisingly finds that Taiwan, the world’s dominant producer of advanced semiconductor chips, is a major source of AI-relevant imports. But under Waugh’s broader umbrella of AI-related goods, Mexico is equally important, with Mexico and Taiwan each supplying about a quarter of U.S. imports. Mexico is a major source for electrical, cooling, and networking products.

Mexico is also a major destination for America’s AI-related exports. Some of this traffic, Waugh believes, represents supply chains that cross the southern border multiple times. AI-related trade also helps explain why U.S. trade with Mexico remained robust in the face of tariffs while imports from Canada dropped. (…)

EARNINGS WATCH

British Airways owner IAG warns Iran war will add €2bn to jet fuel bill Company plans to recoup about 60% of higher costs through savings and raising ticket prices

(…) “If the current conflict continues to restrict flows of both crude oil and jet fuel from the Middle East, there is the potential for supplies of jet fuel to be restricted on a global basis.” (…)

Toyota warns of $4.2bn hit from Middle East war World’s biggest carmaker sells record 10.5mn vehicles last year on strong demand for hybrids

Toyota has said the Middle East conflict will cost it ¥670bn ($4.2bn) in higher component prices and lost sales, becoming the latest carmaker to lay bare the strains caused by the turmoil.

The world’s biggest carmaker said on Friday that the hit from surging prices for parts such as aluminium and rubber tyres, as well as lost sales in the region, would result in a 22 per cent fall in net profit to ¥3tn. That would be the third consecutive annual drop.

“We do not believe we can fully offset negative ¥670bn Middle East impact,” said Takanori Azuma, accounting group chief officer at Toyota.

The surging costs add to the ¥1.4tn burden from US tariffs that was a factor in pushing down net profit by 19 per cent to ¥3.8tn in the 12 months to March, although the company managed to exceed its previous projection.

Toyota’s estimate highlights the growing fallout of the war on the global motor industry after the big three US carmakers sounded the alarm on a $5bn financial hit from commodities inflation. (…)

Azuma said the estimates for higher costs assumed that the war would continue until next March. (…)

The North America region made an operating loss of almost ¥300bn largely because of the tariffs, the company said. (…)

Rule of Law 2, Trump’s Tariffs 0 The Section 122 border taxes go down, his second big legal defeat.

Another tariff swing and another legal miss for President Trump. A 2-1 majority of the U.S. Court of International Trade on Thursday ruled his Section 122 tariffs unlawful. Although the White House may turn to other statutes to dun businesses and consumers, the decision is important for the rule of law and limits on willful presidential discretion.

Mr. Trump invoked Section 122 to reimpose his border taxes after the Supreme Court struck down his emergency tariffs in February. Section 122 lets a President impose tariffs as high as 15% for up to 150 days to address “large and serious balance-of-payments deficits.” Mr. Trump set his rate at 10% across the globe.

The President claims the tariffs are needed to reduce the $1.2 trillion U.S. trade deficit in goods. The legal and semantic problem is that the balance of payments and trade balance aren’t the same, as judges Mark Barnett and Claire Kelly explain. “It is clear that Congress was aware of the differences in the words it chose,” they write. (…)

Mr. Trump’s lawyers argue that the President can still impose tariffs because trade deficits are part of the balance of payments, and the President can pick and choose among the components. “Such an expansive reading of the statute would raise a non-delegation issue, which in turn would prompt a constitutional question,” the judges write.

But the judges say there is no need to address the constitutional arguments since the law doesn’t give the President the authority he claims. “Although the current account (and the balance of trade as a component of the current account) are relevant to balance-of-payments deficits, they are distinct, and the statute recognizes the distinction,” they write. (…)

The judges blocked the tariffs for the business plaintiffs, though they declined to issue a universal injunction.

The practical effect may be minimal since the litigation probably won’t be fully resolved before the law’s 150-day shot-clock for imposing the tariffs ends on July 24. Mr. Trump has also teed up new tariffs under Section 301 against specific countries for allegedly unfair trade practices, which Treasury Secretary Scott Bessent says will soon be unveiled.

(…) it’s hard to recall another President so gung ho about a policy as economically destructive and politically unpopular as are Mr. Trump’s border taxes.

Anthropic weighs deal for near $1tn valuation as revenue surges

Anthropic is weighing raising tens of billions of dollars this summer to fund a vast expansion in computing capacity, in a move that would catapult it past rival OpenAI to a valuation of almost $1tn. (…)

The new round is expected to value Anthropic at about $900bn pre-money and to raise as much as $50bn, said three of the people. They added it was likely to close within two months. OpenAI was valued at $852bn post-money in March after it closed a record funding round of $122bn. (…)

Warming seas are brewing extreme weather in months ahead, scientists forecast Concerns raised about the development of an El Niño warming cycle this year combined with climate change

Sea temperatures around the world were the second highest on record for the month of April, stoking concerns among scientists that an El Niño warming cycle is brewing that would intensify extreme weather.

The naturally occurring El Niño weather phenomenon, where water temperatures in the central and eastern tropical Pacific Ocean become significantly warmer, temporarily accelerates the rise in global air temperature, resulting in the spread of fires, floods and droughts. (…)

The World Meteorological Organization said in March that El Niño had a better than even chance of returning by the end of this year, while in April the US National Oceanic and Atmospheric Administration put the odds of an El Niño returning between May and July at 61 per cent.

At the end of April, the Bureau of Meteorology in Australia said all climate models, including its own, suggested continued ocean warming over the coming month, reaching El Niño thresholds later this year. (…)

“We know that El Niños in general amplify temperatures, so if the impending one is as severe as feared, then we’re in for one hell of a ride,” he said.

“The climate system is complex and predictions are not promises — but I think the coming months and into 2027 are very likely to include a succession of grim environmental stories.” (…)