Strip technology, telecommunications and the energy complex out of the S&P 500 and the remaining seven sectors have grown earnings about 6% annualized this year — roughly 2.5% after inflation.
Aggregate earnings look extraordinary, which is why almost everyone is describing the economy as strong enough to tighten into. Jim Paulsen's charts say the aggregate is two or three separate economies averaged together, and one of them is already in trouble.
"I don't see how you could look at this chart and say that the job market is okay."
Paulsen publishes Paulsen Perspectives, has a recurring show named after him on the Excess Returns network, and brought 27 charts to this one.
The full episode is covered here so you can skip it. 62 minutes of audio, 20 minutes of reading.
Here are the 12 arguments that matter.
👤 Guest: Jim Paulsen, market strategist, who publishes the Paulsen Perspectives newsletter and gives his name to this recurring slot on Excess Returns
🎙️ Hosts: Jack Forehand, who co-founded Excess Returns, and Matt Zeigler, Managing Director at Sunpointe Investments
📰 Published: 14 September 2026 on Excess Returns
🔴 YouTube | 🟣 Apple Podcasts | 🔗 Show notes | ⏱️ 1 hr 2 min | ✅ Time saved: 42 min
Key Takeaways
The S&P 500's earnings boom is three separate stories, and the largest one by company count is barely growing
Technology and telecommunications are just under half the market cap; energy-linked earnings are 5%; the other seven sectors are up about 6% annualized, or 2.5% real
His job-market misery index has only been this high in recessions or in the first months of a recovery
Unemployment at 4.1% minus roughly zero job growth, a reading higher than about 88% of post-war history
Business investment stopped creating jobs a few years ago, and the correlation is now slightly negative
He expects a 4% handle on the 10-year Treasury before 5%, because the surprise index leads the bond market
The wall of worry that supported this bull market is fading, and he treats that as a risk rather than a relief
Economic policy uncertainty has spent the entire bull market in its top quintile, where forward returns annualize at about 20%
AI capital spending has switched from cash-flow financing to debt, and the lowest-rated credit spreads are widening without any stock market reaction
Real profit per job has exploded while output per hour has not, which he calls profit productivity rather than productivity
His forecast is a technology bear market inside a market-level correction, not an S&P 500 bear market
1. Two Economies, One Average
Jack Forehand opened by asking what matters most heading into a Federal Reserve meeting and out of a consumer price report. Paulsen's answer was that the reports themselves are the least interesting part.
He said he pays less attention to the monthly data than most because it creates a day of volatility and then the market moves on, and that what he watches is what policy implies about reports still to come
The framing that runs through the whole episode is that the aggregate hides a split: "But if you look at the separate components, it's really the story of two tales, and one of the tales is really not that good."
The list of pressures he read out is all headed the same way: energy prices spiked back almost to previous highs, the 10-year Treasury yield taken to five, the two-year behaving as though the Fed has already raised, and a yield curve almost back to its flattest of the year
"I mean policy force is very negative and also just the negative lagged force of higher inflation pressure from energy prices bleeding out into everything you know hitting real purchasing power hitting margins and everything," he said
His worry is cumulative rather than event-driven — that the longer negative force is applied to the half of the economy that is already weak, the worse the eventual result
2. A Coin Flip on the Hike
The show put the market odds of a hike at the next meeting at something like 80%. Paulsen would not go along with it.
He called it a coin flip and asked for his coin, and said only that "I think they sure sound like they're going to raise it."
His read of the chairman is that Warsh does not want to raise and feels pinned into demonstrating that he is still focused on inflation. The committee was split at the previous meeting, and he thinks it could be split enough to pass again
Either way he expects the hike itself to be anticlimactic, because the tightening has already happened in the market: a near-5% 10-year, a further inverted curve, and the panic that came with both
The market has already broken below its June high, which was the first high of the AI rally — a technical break he said is not good
What a hold would buy, on his account, is a short-term rally that may not stick, because the real question underneath is the strength of the economy rather than the decision
3. Earnings in Three Colors
The first chart splits forward 12-month earnings per share into three lines, and it is the chart the rest of the episode keeps returning to.
"But below that surface, it's really a two-prong almost three-prong story on earnings here," he said
The blue line is technology and telecommunications, "which I call the brainiac earnings," and is just under 50% of market capitalization
The green line is the companies tied to energy and commodity prices: "And that's what I call the no-brainer earnings because if you're a oil company, you're going to make money." Even a bad manager makes money on that line
The size mismatch is the point: "The blue line's about not quite 50% of the market cap. The green line's only 5%, but it's up unbelievably over this period of time."
The red line is everything else, and it is where the economy actually lives: "If it was just those seven sectors, no one would be saying earnings are great because they're up like 6% annualized year to date." Back out three and a half to four points of inflation and that is about 2.5% real
"No one would be celebrating that as some kind of overwhelming positive force," he said
If he could set policy separately he would tighten on the technology line and ease on the other seven, and his complaint is that no aggregate macro policy can do both — so next week the choice is one or the other for everybody
4. The Job Market Misery Index
Paulsen said the epicenter of weakness is employment, and spent four charts on it.
On the first he is blunt: household employment has been falling for almost 18 months and payrolls have been roughly flat. "The non-farm or household job numbers have been falling for almost 18 months. Not even flat. And payroll has been about flat over that period of time. I don't see how you could look at this chart and say that the job market is okay."
The comfort most people take from low jobless claims is, he argues, misplaced, because claims and payroll growth have parted company for nearly a decade
His objection is social as much as economic: "I think there's a reason why there's so much pessimism on Main Street because a lot of people aren't okay with that." Even someone with a job, he said, has no prospect if no new jobs are being created
He also thinks the causation runs the other way from the consensus: "But one reason I do is because if you look at the last 10 or 15 years, payroll employment has been leading claims, not the other way around."
"I suspect we are going to get some layoffs eventually, particularly if we stay at job growth of zero or worse," he said
The historical test he applies to payroll growth is what makes the section land: "And the only time it's ever been lower is you've been in a recession. We've never had job annual job growth this low without being in a recession." Unemployment, by contrast, is lower than about 80% of post-war history
Rather than choose between them he combines them into what he calls the US job market misery index — the unemployment rate minus annual job growth
"So the misery index at 4.1. And by the way, that's higher than about 88% of the time in postwar history," he said, adding that "the only time it's ever been higher if you look on that chart is when you've been in either in a recession or the first months of a new recovery"
The policy comparison is the sting: on his chart the Federal Reserve has been easing almost every time the misery index has been at this level or worse. "And I'm not sure that's appropriate," he said of a hike next week
5. Capex Stopped Making Jobs
The last jobs chart is the one he seemed least comfortable with, because it breaks a relationship that has held since the war.
He plotted the trailing 40-quarter correlation between annual growth in the investment-to-GDP ratio and annual growth in employment. It has been positive throughout post-war history, strongly so in recent decades
It went slightly negative a few years ago. Business investment, in other words, is no longer creating jobs
He would not claim to know whether that is permanent: he allowed that it might be the new world and that everyone could be fine with it, then said he does not think it is sustainable in this country
The reason it matters now is that business investment is the only genuinely robust part of the economy — which means the one engine running is also the one that no longer employs anybody
6. The 10-Year Could See 4%
Paulsen's bond call runs off Bloomberg's US economic surprise index, restricted to hard data so that nobody can object to the survey component.
"And clearly we've had a slowdown in economic momentum because reports have increasingly now just in the last month and a half or so started to come in worse than expected," he said. The first half of the year ran the other way, which is part of why equities did well
The chart's value is that the surprise index has led the 10-year yield through most of this bull market, and he says very few things lead the bond market
The lag is now at about its normal length, which is why he is on the unfashionable side of the consensus: "I'm more of the view we could see 4% handle on the 10-year again than I am breaking five. We'll see. And that's really a testament to weak economic growth."
7. Retail Without Income
Three charts on the consumer, and the argument is that the spending is not funded by anything durable.
Consumption as a share of GDP tracked labor force participation upward through the whole post-war period and then stopped: "Well, the consumption now has stalled for the better part of the last 20 years in terms of growing bigger. And I think it's because there's less and less labor market participation."
He is watching whether this year's fall in participation shows up as a further drop in consumption's share, and said the historical relationship suggests it would
Real retail sales popped at the start of this year after two flat years, but the support did not follow: "There's been no pickup in real disposable income." Asked whether the pop was just price inflation, he said the series is real and does back inflation out — and that real retail sales have rolled over in the last two months
The three constraints he stacks up are the same ones from the jobs section plus one: "And you think about the savings rate is almost at a record low. And then you come around, there's no real disposable income growth now for the last two and a half years. I think that's going to come home to roost eventually."
8. A Profit Boom, No Economy
Two charts pair the market's internal momentum against the real economy, and both show the same divergence.
He plotted the relative performance of the S&P 500's cyclical sectors — consumer discretionary, financials, materials and industrials — against annual growth in the US coincident economic index. The cyclicals have underperformed badly and fallen further year to date, and on his reading they lead the economy rather than follow it
If the coincident index falls much further, he expects non-farm payroll gains to turn negative, which he said would be hard for anyone to ignore
The bigger divergence is profits. Real corporate profit growth has gone from zero to about 20% in the last year with no corresponding pickup in activity: "We don't get a big profit boom without the economy doing better, except this year we've got this massive profit boom but the economy as a whole is just laid there in the muck."
He raised a measurement doubt as well, on the concentration: "Those companies own each other. A lot of them own each other. And so what you're finding now is part of their profits is coming from how well the other companies have gone up."
"There's little items like that makes you wonder about the quality of the reports that we're getting," he said
His conclusion from the block is a possible change in what the market is frightened of: he said he could see growth becoming more of an issue than inflation in the months ahead
9. The Bull Built on Fear
Forehand asked whether the R-word is on the table. Paulsen would not say it, and then spent the next stretch on why the absence of fear is the real risk.
"I can't even bring myself to do the R word. There's a part of me that says I should," he said, giving three reasons not to: no recession in 16 years outside the pandemic, which he called an exogenous event; a hard call to make; and strong household and corporate balance sheets
The related idea is that the bull market since 2022 has been powered by pessimism: "If you've got everyone worried about end of world scenarios then no one's overexposed. They're all underinvested. They've all got excess liquidity. And they're all waiting for the collapse for the opportunity to buy."
"It's just saying, a bull often runs because of fear. And we've had a massive one," he said
The measure is an economic policy uncertainty index built from newspaper mentions, which he ran back to 1900: "Look at World War I. Look at World War II. Vietnam conflict, all the things we've been through. Nothing's come close to the type of shock and awe we have experienced in here."
He credited the source directly and without much sentiment: "But then we had the Trump phenomena which is just a nothing but a wall of worry creator. It's maybe his best achievement in some regards of this. He created this massive support under the stock market by chronically creating uncertainty and the like."
The whole bull market has sat in the index's top quintile since 1985, and the index is now approaching the boundary. At roughly 175 it is close to the 150 that would drop it into the middle quintiles, and 120 would put it in the bottom two
"Well, from highest quintile, which is our entire bull market, there's an average annualized return of 20% when you're top quintile, which we have been the whole time," he said, and forward returns fall off materially in the middle and lower bands
What is changing is not the news but the response to it: "I think what's happening, we're getting a lot of the same news we've had. It's just that we're becoming more immune to it all." Tariffs and the Iran conflict, he said, no longer get the same coverage or the same reaction
"We're getting close to falling back into normal. That might be a big loss of support for the stock market that people are not at all focused on. People might finally get comfortable for the first time. They might actually feel things are okay."
On sentiment today he drew a distinction between exuberance and complacency: "I say it's just complacent. I mean, when you keep getting told the world's going to end and it never does quite and stocks keep going up and every buy in the dip you do works, I think that's where we're at. It's more complacency. And that's also why a change in the narrative would be huge."
A co-host drew out the inversion: what falls is not the number of things to worry about but how much anyone cares about them
Paulsen agreed, and named the specific flip he thinks would do damage — an audience programmed since the pandemic to worry about inflation suddenly worrying about recession instead
10. Eight Warning Signs
Paulsen listed the familiar warnings quickly — high valuations, the Buffett indicator at a record ratio of market capitalization to GDP, put-call and other sentiment measures, the technical break below the June high, narrow participation — and then spent the rest of the section on the ones he had found in the last month.
The S&P has not yet followed the economic surprise index down, although it has followed it in every previous dip in this cycle. He said he is not sure what level the surprise index reaches before someone notices, but does not think it is far away
Large-cap growth's relative performance has not recovered with the AI rally, and since the dot-com peak an S&P rally has rarely sustained without growth leading it — a 25- or 26-year record he says makes the divergence concerning
He built what he calls a household enthusiasm measure: "And this is just a ratio here of the real purchasing power of wages divided by the unemployment rate." It has been falling since 2023 and 2024 without the usual Wall Street response, and he expects the recent uptick to reverse as oil prices push real wages back down
The ISM services series is the one he called obscure and then made the most of. New orders — which he says are effectively capital spending — have shot up while the employment component has not moved
"One of them was the top of the dot market when that blue line shot up and employment went nowhere over that period of time. And we know what happened to the stock market after that."
"Then again, if you look at 2021 and that bull at the end of that 2021 period, new orders had shot through the roof, but employment ISM services went down. And we know what happened in 2022." What happened in 2022 was a decline of more than 20%
"We've got the same pattern going on today where new orders have shot up and there's been no increase in ISM employment from services sector."
The credit chart is the one that ties the warning to AI directly: "And this is getting increasingly tied to the AI story where tech spending used to be financed almost entirely out of cash flows. Now they're using credit and other means to do it."
The lowest-rated corporate spreads and credit default swap spreads are both widening, and historically any pickup there has brought turbulence in equities
"Now, we've had as big or bigger pickup in credit spreads again just this year and as yet we've had no reaction from the S&P 500," he said
Asked by Zeigler whether a hike would widen spreads further, he said some of it, and explained why spreads have been less sensitive than usual: household debt-to-income ratios have fallen since the financial crisis and corporate cash levels have been good. The change is that the one part of the economy that is growing is now the part borrowing to do it
His historical parallel is the telephone companies before the dot-com bust: "The last part of tech that ever felt that way to me was when the Bell operating companies leading up to fiber optics, they all laid the same fiber optic cable down the railroad beds all across this country." Most of them, he said, did not survive it, and nobody made money on the cable
The asset-allocation chart is the starkest: "And in the last 76 months the stock market has outperformed bonds by more than it ever has over the previous 100 years. Now I picked a specific the specific time period to get that result. I agree with that. But if you use 10 years it's very similar."
Bonds have produced negative returns over that stretch, which has happened only twice in 76-month windows, and this time the negative return is twice the size of the other occasion
His read is not necessarily to sell equities: "5% 10 years might prove to be pretty good for a period of time." Five points of yield plus price appreciation, he said, could compete with stocks, and equities could underperform bonds without collapsing
The last of the set plots gross private domestic investment against the stock market on a log scale, and he said the market is getting ahead of the investment equation. Asked by Forehand why he divides by employment, he explained that what correlates is not the level of investment but how much capital is being applied per worker — labor deepening — and that plain investment would not fit the chart nearly as well
11. Profit Productivity
Forehand asked him to define profit productivity and explain why he watches it instead of the official measure. The answer is the analytical core of the episode.
Measured productivity is up a little, he said, but weak growth flatters it: in a slowdown companies cut jobs before sales fall, so output per worker rises for reasons nobody should want
The 1990s were the real thing: "We had rising output with rising jobs going on and still had rising output per job. So we had like two or 3% 3% job growth and we had you know two or 3% productivity growth all at the same time. That's not what we got here. We got maybe 2% productivity, but that's with jobs, going to zero essentially."
What has exploded instead is real profit per job, which he has tracked against price-to-earnings multiples and says is part of why valuation ranges have been permanently higher since the 1990s. Before the 1990s the series was roughly flat all the way back to the war
The distributional consequence is explicit: profit margins are at record highs at the same time as labor's share of GDP is at record lows. "That's because profit productivity is making up the difference in a big way," he said
The reason he is worried about it now is a lead-lag chart: the 10-year-to-two-year yield curve, shifted forward four quarters, has historically led the profit-per-job cycle, and the curve is heading back toward a new low. "You got two-year yields going up a lot faster now than the 10-year yield," he said
On that timing the peak in the leading line arrives around the end of this year, and a roll in profit per job is the single thing he thinks would change the narrative most, particularly in the technology sector
12. A 20% Tech Bear Market
The final section goes back to the three-color earnings chart and works through each line forward.
The red line — the other seven sectors — he expects to get worse rather than better: "Real money growth's around 1 and a half to 2% in the last year, which is pathetically low. There's not enough room even to grow GDP much more than 2%." Higher oil prices are eating purchasing power and policy is tightening on both the monetary and fiscal side
The green line comes down on arithmetic rather than on prices falling, because growth rates are what feed earnings: "That is oil prices may still be high, but they're no longer rising on a growth basis. And that's what matters for profitability. It's not the level because costs catch up. It's they got to keep going up." Oil has been flat since March, so the comparison rolls over in a quarter or two
That leaves the blue line as the wild card, and he doubts it can keep the pace even if nothing goes wrong, because it started from such a low base this year
He also flagged the financing shift inside it: the ratio of corporate cash flow to new-era investment spending has come down sharply in the last two quarters, and historically the end of cash-flow-funded technology spending has ended technology cycles
His forecast, stated plainly: he is not calling a bear market for the S&P 500, but he does expect one in technology — a decline of more than 20% in the technology and telecommunications companies. The other seven sectors go down somewhat and, having not risen much, he thinks largely hold up, which would produce a 10% to 15% correction at the index level
He does not treat that as a disaster: "But it would also bring policy easing and some of the things I think the broader economy needs here for this thing to continue if you will."
The warning he closes on is about letting the split run: "If we carry this on too long, I think we do create a situation where the divergence between profitability and jobs in this country becomes so extreme it caves in on itself at least for a period."
Bonus Insights
Forehand counted 26 charts at the top of the show and teased Paulsen for slacking off from 34 last time; Paulsen corrected him at the end to 27 and said he had short-changed himself
Paulsen publishes at paulsenperspectives.substack.com, puts out about two pieces a week, and said he is too old to do more than that
He lives in Minnesota, where, he said, everyone spells the name with an O — he is an import and spells his with an E
Forehand's sign-off arithmetic was two pieces a week and 27 charts an hour
Paulsen's bottom line is that the strongest earnings cycle in years is concentrated in a sliver of the market that is now borrowing to fund itself, while the job market underneath it is as weak as it has ever been outside a recession — and that a technology bear market, a 10% to 15% index correction and the policy easing that would follow are the likely way that resolves.
Products, Companies & Tools Mentioned
Bloomberg (The source of the US economic surprise index Paulsen uses in the hard-data-only form, which he says leads the 10-year Treasury yield through this bull market)
Institute for Supply Management (Its services report supplies the new orders and employment series behind the dot-com and 2021 comparisons)
Federal Reserve (The meeting he calls a coin flip; his chart shows it easing almost every previous time his misery index was at this level)
Books & Resources Mentioned
Paulsen Perspectives – Jim Paulsen (Where all 27 charts and about two pieces a week are published)
If this was worth your time, send it to someone who has to have a view on this.
Get the latest market chatter as it happens: