If the first half of 2026 teaches us anything, it’s that worrying about geopolitics doesn’t make you rich.
So far this year we’ve seen President Trump kidnap Maduro from Venezuela, threaten to invade Greenland, launch an actual real-life war against Iran that shut the Strait of Hormuz. Oh, and he might even have pulled off a successful coup at the Fed.
The US market powered on through, making its investors around 8 per cent richer so far this year. So what’s driving things higher? Well . . .
As we wrote back in April, this hasn’t been the case of fingers-in-ears-I’m-not-listening-and-will-buy-stocks-at-any-price. The rally has been accompanied by a huge uplift in analyst earnings expectations. And compared to aggregate year-ahead analyst earnings forecasts, US stocks are actually a little cheaper than they were at the end of last year.
Some content could not load. Check your internet connection or browser settings.
In fact, it looks like analysts were at the start of the year forecasting total adjusted earnings of just over $2.5tn for the year ahead for MSCI USA constituents. With almost half the year gone, we’re not looking at the same period, but analysts now reckon the same 525 companies should earn a collective $3tn over the next year. This is a huge change.
We thought it would be fun to test the limits of dataviz to slice and dice these figures and walk through just how we’ve got here.
Looking at the sector level, energy is the standout winner.
Some content could not load. Check your internet connection or browser settings.
By our count, at the end of 2025, analysts reckoned that total year-ahead earnings of all the 22 energy companies in the MSCI USA index would be $103bn.
That’s pretty understandable. If soaring oil prices are going to help anyone, it should be energy companies. By May spot crude oil prices had almost doubled and year-ahead analyst earnings expectations were over $150bn. Since then crude oil has collapsed back to close to prewar levels. So far expected energy earnings haven’t really followed spot crude back down.
This all obviously matters a lot to investors in energy companies. But does it matter much to folks who just hold the index? A bit.
Two of the biggest contributors to the overall increase in expected US market earnings come from Exxon and Chevron, which between them contribute more than half of the uplift in energy sector earnings expectations.
But they’re dwarfed by earnings upgrades across large-cap tech as we hope this (possibly very Marmite-ish) chart shows:
Some content could not load. Check your internet connection or browser settings.
We appreciate that not every Alphaville reader is a Marimekko chart convert. So let’s take it step by step.
We know that US large-cap stock indices skew to technology. This is partly because tech companies make a ton of money, and partly because people pay through the nose for them, inflating their market capitalisations. Intel, for example, is expected to make around $5.8bn over the next year, but is valued at over $580bn. ExxonMobil is also valued at around $580bn, but it is expected to earn over $45bn over the same period.
Given we’re looking at earnings, not market cap, we’ve sized the bars on the vertical axis for the 11 industry categories by total expected earnings contribution. While less dominant than an equivalent chart showing market cap, tech is the largest source of expected earnings, bringing around $1tn to the party. So it’s the thickest slice.
Meanwhile, the horizontal axis shows how overall sector earnings expectations for the year ahead have changed since the beginning of the year. Just like our line chart, energy comes out on top, followed by tech. But the importance of tech jumps out by virtue of its earnings size.
To make the chart even cooler, we’ve picked out the top 10 individual companies’ contribution to this year-to-date uplift in individual sector expected earnings. You need to hover your pointer over the coloured blocks to see what they are because whacking them on a legend made the whole thing unreadable.
A couple of things jump out to us.
The first is Micron, in salmon pink, on the tech bar. By our calculation it provides a bigger contribution than Nvidia to the overall uplift in US large-cap earnings expectations. Maybe this is just what happens at this stage in the cycle. And perhaps, given its latest blowout quarterly earnings, this won’t surprise everyone. Still, not bad for what was, not long ago, a little Boise-based semiconductor maker.
The second is the degree to which the overall move higher in index-level earnings expectations derive from upgrades to just a few companies. Sure, lack of breadth in US markets is nothing new. But even the heady market cap weightings of US stock indices of the biggest companies is dwarfed by the two-thirds of the overall boost in earnings expectations this year that are expected to come from just 10 companies:
Some content could not load. Check your internet connection or browser settings.
But how have these 10 stocks performed for investors? Variably.
Micron has been positively South Korean — roughly tripling in value. But the collective market cap of the other nine companies has fallen by around 2 per cent since the start of the year, underperforming the market as a whole.
Sure, there’s a case for disregarding forward-looking analyst earnings as systematically Panglossian. But the alternatives don’t look great.
GAAP numbers are both backward-looking and are subject to being swung around by unrealised revaluation gains on stakes in private companies (like OpenAI and Anthropic). As Owen Lamont, a portfolio manager at Acadian, points out, the ASU 2016-01 accounting rule on unrealised capital gains can lull investors into seeing rising earnings as a justification for rising prices and therefore cause a mechanical feedback loop.
We know that individual stocks won’t always, or even often, closely track the short-term views of analysts. It would be weird if they did.*
But it is a little odd to see the dominant market narrative being of a US earnings-fuelled rally where the companies doing the earnings-fuelling aren’t really the companies doing the rallying.
*Equities being, after all, the present value of every year’s earnings from here to infinity. Also, given that we’re looking at how blended 12 month forward consensus earnings are developing over a six-month period, it would be possible that the subset of companies delivering the greatest uplift were expected to deliver more (although they weren’t). We thought about stretching out the exercise to examine analysts’ two- and three-year-ahead estimates to improve things, but bitter experience has taught us that these tend to aggregate to next-year-plus-ten-per-cent come rain or shine. Long-term residual growth rate forecasts are even worse.
Further reading:
— Earnings forecasts up, stocks . . . down (FTAV)
— The mysterious $53bn ‘other income’ boost to AI hyperscaler earnings (FTAV)
