Why Wall Street Bulls Aren’t Worried About Sky-High Stock Prices Net profit margins are higher than average across multiple sectors, indicating corporate America has grown more resilient

It’s the metric giving Wall Street comfort over stocks’ sky-high valuations.

A historic surge in corporate profits has helped lift a key indicator of underlying business health to a record high, giving some investors reassurance about the sustainability of the market’s runaway gains.

The net profit margin for companies in the S&P 500 rose to 14.8% in the first quarter, according to FactSet. This marks the highest net margin, a measure of the profit generated from every dollar of revenue, reported by the index since the data provider began tracking this metric in 2009. The previous peak of 13.2% was set just a quarter earlier.

It isn’t just tech companies, either. In the first quarter, multiple sectors including financial services and industrials reported net margins above their five-year averages. For investors, the broad-based strength suggests corporate America has gotten more resilient to geopolitical conflicts that could trigger an inflationary jump and economic slowdown.

“This is a productivity‑driven environment, much like the 90s were, and productivity is spreading across sectors,” said Nancy Tengler, chief executive of Laffer Tengler Investments. “It’s thanks to not just AI, but all the new technologies.”

While the margin expansion is broad-based, the tech sector still drives the bulk of recent growth, fueled by companies such as Nvidia and Micron Technology. Excluding the tech sector, the S&P 500 would have reported a net profit margin of 12.4% for the first quarter, according to John Butters, senior earnings analyst at FactSet. (…)

The earnings growth rate for companies in the S&P 500 surged to 28.8% in the first quarter, the highest level since the fourth quarter of 2021.

Already, analysts are predicting a robust second quarter, which could serve as a key test of how profit margins held up against the inflationary pressures stemming from the conflict in the Middle East and heavy spending on the AI infrastructure boom. (…)

A key concern is the sustainability of the expanding margins. Because the technology sector has driven outsize growth, any reversal in its pricing power or demand could cause earnings to decline rapidly. OpenAI, for instance, is considering slashing the prices it charges users to win customers from rival Anthropic, the Journal reported.

“The problem is this can whipsaw pretty fast, and if the dynamics change on pricing, on the insatiable demand for semiconductors, in particular, this reversal could be significant,” said Matt Miskin, co-chief investment strategist at Manulife John Hancock Investments.

Valuations also remain a concern, though robust corporate earnings have so far justified elevated stock prices. The S&P 500 is trading at about 20 times its projected earnings over the next 12 months, higher than the 10-year average of 19, according to FactSet.

Profit margins could also face pressure from tighter financial conditions and elevated interest rates, which could translate into higher borrowing costs across the economy. (…)

Several warnings here:

1- The WSJ reporter, like most pundits, fails to mention that Q1 earnings were boosted by hyperscalers’ mark-to-market gains from equity stakes in private companies. “Alphabet and Amazon generated “other income” totaling $53 billion in Q1 2026, which accounted for nearly 60% of those two companies’ income in Q1 and 34% of the total $155 billion in income this quarter across the five largest hyperscalers.”  (GS)

These huge, unusual, non-operating profits boosted the S&P 500 YoY earnings growth by about 12%, from 16% to 28%.

For some “strange” reason, there is little effort to publish normalized earnings, even though Q1 earnings are meaningfully distorted and will serve as an inflated base for future earnings.

And that is before upcoming earnings results which will need to mark-to-market the continued appreciation in private AI companies during Q2 and Q3. Particularly GOOG, NVDA and CSCO.

Rough calculations suggest that 2026 S&P 500 EPS are currently inflated by about $20 and its P/E ratio by about 1.3x.

2- Normalized profits are still up a spectacular 16%, but the bulk of the growth comes from hyperscalers. Goldman Sachs calculates that ex-AI, S&P 500 earnings are not being upgraded and that median earnings growth will be around 9% in coming quarters.

 

As I wrote yesterday, only 4 sectors (Energy, Materials, IT and Comm. Services) are growing earnings faster than the S&P 500 average this year. The other 7: +9.1% on average. For 2027, only 3 sectors (Health Care, Industrials and IT) should be growing earnings faster than the S&P 500 average. The other 8: +7.3% on average.

3- Obviously, failing to normalize profits results in large distortions to profit margins. True, margins have also increased outside of IT but not that much as Ed Yardeni illustrates:

4- In the wake of the pandemic and the wars in Ukraine and the Middle East, consumers accepted price increases as inevitable and justified, particularly since their own incomes were also rising amid a rather strong economy. That’s no longer true. Real income from labor or total disposable income have been essentially flat YoY since March.

The recent Fed Beige book carried many corporate statements on the changing landscape:

• “Output prices rose slightly on average, although many firms left prices unchanged. Input prices increased modestly overall, but some contacts reported significant cost pressures.”
• “Most contacts did not plan to raise their output prices in the near term, even though many were concerned that cost pressures linked to the Middle East conflict could persist for a while.”
• “Firms continued to report moderate increases in prices received for their own goods and services. However, a dichotomy has emerged. Increases in prices received by consumer-facing firms have held steady at about 2.0 percent for the past six months.”
• “Most contacts across many sectors reported a reluctance to raise prices, citing consumer price sensitivity and softening demand.”

5- While on costs vs selling prices, the WSJ is right questioning the sustainability of tech margins. “Because the technology sector has driven outsize growth, any reversal in its pricing power or demand could cause earnings to decline rapidly. OpenAI, for instance, is considering slashing the prices it charges users to win customers from rival Anthropic, the Journal reported.”

But now consider Chinese vs Western LLM pricing. Chinese providers have turned LLM pricing into an explicit competitive weapon, cutting API prices multiple times since 2025 and capturing a majority of token volume on aggregators like OpenRouter.

In case you question Chinese LLM quality. At the very far right, Zhipu’s LLM (Z.ai) performance has almost reached Anthropic’s best model, at a fraction of the cost.

Apollo’s Torsten Slok:

The first chart below shows that so far there are no signs of profit margins rising outside the tech sector. This is ultimately what we are waiting for, because the value of AI companies today rests entirely on the promise that margins in the S&P 493 will eventually climb.

That promise is the link to current market prices, since implicit in the valuations of AI companies are assumptions about future earnings. That’s why the current debate about token costs, model routing and token marketplaces is important. If token costs converge toward zero for most AI use cases, then there is not enough revenue for all hyperscalers even in a situation where compute demand surges higher.

The key issue is the length of the ROI runway outside the tech sector. In a handful of sectors, software and tech above all, implementation is nearly immediate, since these firms can fold AI into their own products and processes overnight.

But that is the exception. Across most of the economy, and especially in capital-intensive, heavily regulated sectors, deep process re-engineering and data governance requirements could delay structural productivity gains well beyond what the market currently projects.

The list of slow-moving sectors is long, spanning health care, banking and insurance, energy and utilities, defense and aerospace, pharma and life sciences, manufacturing, transportation and logistics, construction and real estate, education, legal and the public sector.

This creates a dangerous divergence between aggressive, front-loaded valuations today and a much slower cash flow reality, since equity markets priced for instant earnings growth will face a painful repricing if the productivity hockey-stick takes five years rather than five months, see the second chart below.

Put differently, companies will slow their AI spending if they don’t see ROI quickly, and the current focus on token optimization is an early warning that AI implementation could be a bumpier, slower road than expected.

The bottom line is that a mismatch between current earnings expectations and the actual time firms need to generate ROI on AI investments could have significant implications for many AI company valuations today.

Beware of extrapolations, particularly on unusually inflated numbers.

AFFORDABLE LEVERAGE?

Karoline Leavitt, the White House press secretary, described it as “one of the most significant pieces of housing affordability legislation in American history,” adding that the “historic bill signing is another promise made, promise kept.”

But Mr. Trump sharply reversed course on Wednesday morning. In a series of social media posts, he said the housing bill was “of minor importance” to other priorities.

Democrats and Republicans began the day on a jubilant note, as they prepared to commemorate a new set of policies meant to lower housing costs — a long-sought priority that both parties hoped to trumpet to voters anxious about the economy ahead of the midterm elections.

But Mr. Trump swiftly and unexpectedly upended those plans. Hours after his own aides praised the bill and promised the president would sign it, Mr. Trump instead canceled a scheduled event at the Capitol. Eschewing an opportunity for rare bipartisan accord, the president opted to turn the bill into political leverage, aiming to force Congress — and members of his own party — to bow to unrelated demands over voting restrictions and the war with Iran.

Mr. Trump insisted on social media that he would only sign the housing measure into law if Congress first approved the SAVE America Act, a divisive measure that would impose new requirements for voter identification and limits on mail-in ballots. Republicans have been unable to advance that bill over fierce Democratic opposition. (…)

The White House declined to say if the president might sign the bill another time, or if he might take the more extreme step of vetoing a law that his party views as a way to convince voters that Republicans are doing something to address the cost of living. (Axios)

The inflation conundrum

During his February 16, 2005, Congressional testimony, then-Federal Reserve Chairman Alan Greenspan coined the term “conundrum” to qualify the decline of long-term interest rates while the Fed was actively hiking.

This economic anomaly seemingly decoupled the Fed’s traditional monetary policy from the broader bond market, confusing policymakers and economists alike.

The current unusual decoupling of core inflation from trimmed-mean inflation, Kevin Warsh’s preferred inflation indicator, is also confusing and is fueling a critical debate about what is the “true” core inflation rate?

Trimmed mean inflation is a measure of inflation that drops the biggest price movers (both spikes and plunges) before averaging the rest.

For the Dallas Fed trimmed-mean inflation rate, 24% of the weight from the lower tail and 31% of the weight in the upper tail are trimmed. “Those proportions have been chosen, based on historical data, to give the best fit between the trimmed mean inflation rate and proxies for the true core PCE inflation rate.”

Warsh’s preferred indicator was designed and weighted to closely mimic the indicator most people watch!

Between 2010 and 2020, trimmed-mean PCE inflation ran slightly higher than core PCE inflation. Not only has it been lower since, but, since October 2026, trimmed-mean inflation has slowed from 3.4% YoY to 2.4% while core PCE inflation accelerated from 3.4% to 4.3%.

Many economists conclude that, since the “belly” of inflation is slowing, the true trend is down.

Mechanically, when tails are small and symmetric, trimming has limited effect; the trimmed mean is then primarily the average of the “middle” categories, which were often running a touch hotter than aggregate headline because some goods disinflation (China, tech) pulled headline down like in the 2010-20 period.

The post‑pandemic period looks structurally different:

• Price‑change distributions for PCE became strongly right‑skewed: a relatively small set of categories (durable goods like new/used cars, certain services, fees) experienced very large increases, while many others moved more modestly.

• The Dallas Fed procedure trims off a large fraction of the upper tail; with skewness, that means it systematically removes more upward pressure than downward pressure, biasing the trimmed mean downward relative to the true underlying trend.

Cleveland Fed work quantifies this: when the cross‑section of price changes turns more positively skewed—as it did after 2021—both median and trimmed‑mean inflation show a time‑varying bias, typically understating trend by tens of basis points unless adjusted for skewness. That’s why since roughly 2021–22, trimmed‑mean PCE tends to run meaningfully below core PCE and often well below headline.

Why the sharp divergence since last October:

This informative Dallas Fed chart “plots the evolution of the distribution of price increases in the monthly component data over the past year. The chart shows the percentage of components each month, weighted by their shares in total spending, for which prices grew between 0 and 2 percent (at an annual rate); between 2 and 3 percent; between 3 and 5 percent; between 5 and 10 percent; and more than 10 percent.” It illustrates what percentage of consumer spending is experiencing low, moderate, or extreme price increases.

The vertical axis (Percent of Expenditure) shows the cumulative share of the total PCE consumption basket.

The colored bands only stack up to roughly 70–80% of total expenditure. The white space remaining at the top (my black rectangle) represents components experiencing negative price growth (deflation, or less than 0%).

The bands segment price growth by severity, ranging from optimal/low inflation (Purple: 0–2%) to extreme inflation (Green: More than 10%).

The Dallas Fed automatically cuts the lowest 24% of price changes. Looking at the chart, this means the white deflationary zone at the top and a sliver of the purple (0–2%) band are systematically discarded every month.

The Fed also cuts the highest 31% of price changes. This means the entire green band (More than 10%), the entire blue band (5–10%), and a portion of the orange band (3–5%) are stripped away.

The Trimmed Mean inflation rate is calculated only from the remaining middle section—predominantly the cyan (2–3%) band and most of the orange and purple bands.

We can observe that, since November 2025, a growing share of spending is in categories with >3% annualized monthly inflation, but many of the very fastest‑growing categories lie in the top 31% of the distribution and get trimmed out.

Recent research from the Dallas and Richmond Feds argues that the fixed 24/31% cutoffs can produce downward bias when positive skew and global supply‑side shocks dominate, exactly like the current environment.

So, since last October, headline/core PCE are being pulled up by a relatively narrow cluster of high‑inflation components, while the trimmed‑mean algorithm is stripping much of that out and averaging a middle that has not moved as much. That produces a falling or flat trimmed mean against a rising core/headline.

Without adjustment, trimmed mean is now a biased low estimate of trend, especially in periods when skewness spikes.

In reality, inside the trimmed regions (the belly), a lot of categories have moved from “moderately hot (2-3%)” to “hot (3-5%)” so the belly has actually fattened as my red arrows show, with a larger part of it being 3-5% inflating, which has not (yet) been reflected in the trimmed-mean series.

The Cleveland Fed calculates its own 16 percent trimmed-mean CPI, “a weighted average of one-month inflation rates of components whose expenditure weights fall below the 92nd percentile and above the 8th percentile of price changes.” A symmetric and substantially less radical trim. The May data show Core, Median and Trimmed-Mean inflation all at 2.9% and rising.

Kevin Warsh is closely watching an indicator that two Federal Reserve Banks say is flawed, currently downwardly biased by global supply-side shocks, and thus not reflecting the reality. An indicator that for 10 years prior was actually consistently somewhat higher than core inflation.

Mr. Warsh is way too smart not to know this …

Another Federal Reserve bank, The Richmond Fed, found that a panel of 530 CFOs ranging “from small operations to Fortune 500 companies across all major industries” said that their most pressing concern was inflation (25% in Q2 vs 9.5% in Q1), particularly non-labor costs.

CFOs are seeing inflation in real time which Ed Yardeni illustrates with these charts:

Hopefully, ceasefires, a mild labor market and potential AI-productivity will combine to quickly stop the build-up in price pressures throughout the pipeline. Yes, everything energy-sensitive will deflate in coming months, but watch the belly.

Because inflation matters, more than most investors think. Whatever his bias, Kevin Warsh was quite clear 2 weeks ago: “This committee will deliver!”, even though the almost same committee has not delivered for 5 years.

It matters particularly when valuations are very stretched …

@GlobalMktObserv

… on profits boosted by spiking margins …

… leading analysts to project LT profit growing 25.5%, even faster than 2000’s 19.5% expected growth rate (actual was 5.5%).

Analysts currently forecast S&P 500 earnings rising 26.6%, slowing to +17.4% in 2027 and to 13.1% in 2028 (I know, doesn’t jibe with the 25.5% LTEG above, but still 18.6% CAGR, the very high end of history).

FYI, only 4 sectors (Energy, Materials, IT and Comm. Services) are growing earnings faster than the S&P 500 average this year. The other 7: +9.1% on average.

For 2027, only 3 sectors (Health Care, Industrials and IT) should be growing earnings faster than the S&P 500 average. The other 8: +7.3% on average.

Inflation also matters when leverage is unsuspectedly high.

• The Trillion-Dollar Borrowing Binge Lifting the Stock Market to Risky Heights Leveraged funds and margin debt have grown to unprecedented levels this year

Investors have never been more eager to ratchet up their stock returns through margin loans and funds that amplify gains and losses. It may be a sign of trouble. 

U.S. margin debt, or what investors borrow from their brokerages to buy securities, rose 54% to a record $1.4 trillion in May from a year earlier, according to Finra data. Meanwhile, high-risk leveraged exchange-traded funds that produce double or triple the daily move of underlying stocks are growing rapidly, as is trading in options tied to them.

The risks surfaced this past week in South Korea, a market dominated by highflying semiconductor stocks and rife with investors eager to pile on leverage. Korean stocks gyrated, triggering circuit-breakers meant to stop losses on the way down. As the souring mood spilled into U.S. trading, hitting AI-related stocks, a chorus of investors and analysts sounded the alarm that leverage was building up here, too.

“I’m fearful that we’re building unintended leverage that isn’t fully understood,” said Mark Hackett, chief market strategist for Nationwide’s investment management group. “You’ve got people with a lottery mentality using margin to buy options on levered ETFs. That’s three or four layers.”

Buyers ranging from hedge funds to teenagers on Robinhood have poured money into leveraged ETFs this year, helping to nearly double the assets in these funds to a record $220 billion between March 30 and June 3, according to FactSet. (…)

The risks of buying leveraged funds are well-advertised: a 30% drawdown in the underlying stock can turn into a 90% wipeout for the fund. But Wall Street sees a broader problem emerging: These funds, along with other forms of leverage, can also affect how the individual stocks behave. The recent action in South Korea, market participants said, offers a glimpse at the risks.

In a bid to keep pace with the flow of new money, leveraged funds have bought some $300 billion in derivatives linked to single stocks and indexes since the end of March, Barclays analysts estimate.

Those purchases have in turn spurred demand for underlying shares from market makers, which buy stocks to hedge their exposure to the derivative contracts they write. 

That’s almost certainly contributed to the sharp gains in corners of the stock market this year. But when stocks fall, leveraged funds lose assets. That forces them to reduce exposure to the shares they track, which in turn threatens to pull down stock prices even more. There is a danger, market analysts said, that this cycle can quickly spiral into heavy losses.

“That’s a somewhat terrifying figure to contend with should it need to be unwound in a short period of time,” Alexander Altmann, global head of equities tactical strategies at Barclays, told clients on Wednesday. “This is without a doubt the largest nondiscretionary driver of risk at the moment.” (…)

“I am genuinely worried we have so much money going into the complex of leveraged single-stock products because the more money that goes in there, the more this procyclical trading effect happens,” said Dave Nadig, an industry veteran and director of research at ETF.com. “Anytime you have any known-in-advance, price-indiscriminate buyers and sellers, you have a problem.” (…)

South Korea’s top financial regulator said earlier last week that he regretted not blocking the launch of leveraged single-stock funds.

“These are high-risk products, and it seems like about 92% of holders are retail investors,” Financial Supervisory Service Gov. Lee Chan-jin said at a press briefing. “Despite consumer warnings, trading hasn’t cooled.” (…)

This can explain some of that:

The Lag-7 (Callum Thomas)

The divergence widens —Mag-7 is increasingly being rebranded as the Lag-7 as Mag-7 dropped -14% from the 14th May peak (vs ex-Mag-7 up +5% over the same period). But note that tick-up from oversold levels on Friday.

@neilsethinew

• FYI

(Koyfin)

• SpaceX set to join Nasdaq 100, paving way for wave of passive buying July 7

“We the People”

Bruce Mehlman writes that

402 of 435 House seats are not competitive. While the nation is almost evenly divided, fewer than 10% of Congressional seats are close. Thanks to aggressive gerrymandering, geographic self-sorting and winner-take-all general elections, the winners of partisan primaries are all-but guaranteed to capture >90% of seats. What happens to a system designed for competition-driven-compromise when there’s so little competition?

But is the nation so evenly divided? K-shaped politics!

Yes, the number of Democrats and Republicans the same at 27% but declared “Independents” are now 45%, up from 35% 20 years ago.

The Constitution says nothing about political parties, which are neither required nor preordained. Yet the Republican & Democratic parties have dominated U.S. politics for most of the past two centuries.

Registration data suggest voters are increasingly frustrated by the status quo but unable to change much given rules entrenching the duopoly (partisan primaries; winner-take-all districts).

If 45% of voters don’t want to be Democrats or Republicans, should 99% of Representatives be Democrats & Republicans?

Money rules America.

Campaign cost are increasing much faster than inflation, with the richest Americans increasingly dominating election finance. Does this make our democracy more robust & resilient, or does it undermine trust, accelerate alienation & drive populism?

In my December 23, 2025 post Ten Days In China: Who got a plan?, I discussed China’s obsession with long-term planning:

• Five-Year Plans provide 5-10 year stability. Goals persist across leadership changes, allowing for long-term R&D investments.
• The US approach historically avoids broad, “socialistic state-led” industrial planning. Instead, it uses the capitalistic “market-correction” mechanisms to nudge the private sector.
• The government sets broad goals and offers incentives for private companies to figure out the solution. It relies on private competition to determine winners.
• US policies are generally legislated or, increasingly, set by presidential decrees that can be altered by future administrations. They lack long-term guarantees and in the case of decrees are often reactive rather than thoroughly planned, analyzed and discussed.

Bruce Mehlman again: Can America Plan Long-Term When the Rules Keep Changing?

Every two years the partisan alignment in Washington changes. Republicans cut Democratic projects when they’re in charge, while Democrats end GOP initiatives when they’re in control. Congress hasn’t funded the government on time since 1997, governing mostly by short-term stopgaps…

Administrations operate via regulations and executive orders that get tossed by courts & successors. How can businesses (or federal agencies) plan & invest for the long-term when rules & incentives keep flip-flopping?

Europe’s Heat Wave Is the Worst on Record, Researchers Find

Researchers at World Weather Attribution, who looked at heat and humidity levels during both the day and night during three consecutive days in the month of June, found that temperatures were between 5C and 12C above the seasonal averages across France, Germany, Italy, Spain and southern England. (…)

“Yes, this is climate change,” said Friederike Otto, a professor at Imperial College London and a co-author of the study. “Yes, it’s us, no, it’s not El Nino, yes, we have the solutions, no, we’re not implementing them fast enough.”

The current heat wave is now shaping a “structural investment trend,” Sarah Kapnick, JPMorgan Chase & Co.’s global head of climate advisory, said in an interview with Bloomberg Television on Friday. “The stresses that we see today are only going to get worse because heatwaves like this ten years from now will be over 40C and it will keep going.” (…)

• Americans Want to Apologize for Doubting Europe’s Heat: ‘I Did Get Humbled’

Lucy Kloc, a Floridian living in the U.K., used to make fun of her British boyfriend for complaining about the summer heat. How bad could it be, the Jacksonville native thought? She’s endured humid, 90-degree-plus summers most of her life.

“Before, I was like ‘This is nothing, you’re just being a weenie about it,’” said Kloc, who moved to Manchester last year. Now sweating her way through a record-breaking heat wave, Kloc is ready to admit that she was wrong.

“I really did think they were just being dramatic, that it can’t be that bad,” she said. “But it is.”

Americans in Europe have been getting the shock of their lives this summer as temperatures soar to new records across the continent. (…)

“I did get humbled,” said Marissa Parks, a 30-year-old nurse who moved to London last year from Houston. “Being a Texan, it feels kind of embarrassing to be in London and be like, ‘I can’t handle the heat here.’” (…)

“These heat waves hit different,” she said. “You can’t get any reprieve from it. Once you do get hot, you stay hot.” (…)