0
0
You can put in the same hours you put in five years ago, do the work at least as well, and still feel like less of it reaches your table.
That feeling now has a number attached to it. In the second quarter of 2026, the labor share of income in the United States — the slice of everything the economy produced that went to the people who did the work — fell to 52.8% of GDP. That is the lowest it has been since the government began keeping this particular record in 1947.
In the same three months, corporate profit margins hit a record high of 14.9% of GDP.
Both figures were flagged by Gregory Daco, chief economist at EY-Parthenon, and reported by Fortune on 4 September 2026.
What “labor share of income” actually measures
It sounds technical. It isn’t.
Every year a country produces a certain amount of value — goods made, services done, food grown, code written. That value gets divided. Some goes to the people who did the work, as wages, salaries and benefits. Some goes to the people who own the business the work was done for, as profit.
The labor share of income is just the first slice, written as a percentage.
And here is the part worth holding onto: it is not a measure of how hard anyone is working. It is a measure of where the result lands.
The record keeps breaking, and that is the real story
Here is what didn’t make the headline.
This is not the first time in 2026 that this record has been reported broken. The same finding — lowest labor share since the series began — was written up in January. It appeared again in May. Again in August. Now again in September.
Every one of those reports was accurate. That is exactly what makes it worth noticing. A single bad quarter is a spike, and spikes correct themselves. What this past year looks like instead is a floor that keeps quietly giving way.
Daco is direct about that. He does not treat 50% as some natural stopping point. “I don’t think there’s a floor,” he wrote in his note to Fortune.
For a sense of scale: reporting on the same long-run series puts labor’s share at around 70% back when the record started in 1947.
It isn’t AI — at least, not yet
Almost every version of this story reaches for artificial intelligence. It’s the obvious explanation, and it’s the one that travels.
Daco’s own explanation is different. He attributes the gap mainly to productivity gains from automation and cost-cutting, concentrated in large, capital-rich firms rather than spread around — and he points out that the AI boom hasn’t really arrived yet.
Read that twice, because it turns the story around. The number got here without AI.
Which isn’t reassuring. It’s the opposite. If the split moved this far on ordinary automation and ordinary cost-cutting, then AI isn’t the cause of what has already happened — it’s the next chapter. That fits what the people running these companies say when they’re asked plainly: when nearly 6,000 executives were surveyed about AI productivity gains, nine in ten reported none so far, and then predicted them anyway.
What a national number changes for one person
Honestly? Nothing about your Tuesday. A statistic doesn’t pay anyone’s rent, and no chart has ever made a tight week less tight.
But it does change one thing, and it isn’t small.
If you have been quietly assuming that things feel harder because you’re managing badly, or not trying hard enough, or missed some move that everyone else apparently made — the numbers say the ground moved. That isn’t an excuse for anything. It’s just an accurate picture. And an accurate picture is a much better thing to stand on than a private theory that you are the problem.
It also explains something a lot of people are feeling and not saying out loud. When the split tilts and hiring goes quiet at the same time, staying put stops feeling like loyalty and starts feeling like the only safe move available. That’s the pattern underneath the rise of so-called “job hugging” — people holding on to roles out of fear rather than choice.
One more thing worth saying plainly: these are American figures, measured in an American way. But the feeling underneath them — doing the work and watching less of it arrive at your own table — is not confined to one country. That part travels everywhere.
And there’s something odd about how old this particular problem is.
The earliest written instructions we have about work aren’t about growth, or efficiency, or output at all. They’re almost entirely about the person at the end of the chain. One of them says to pay a laborer before the sun goes down — not at the end of the month, not when it’s convenient — for the plain reason that the person is counting on that money tonight. Another says that when what a worker earned doesn’t reach them, it makes a sound. Not as a figure of speech. A sound, that gets heard.
Whoever wrote that down assumed God was paying attention to the arithmetic. Not to the national average — to the one specific person doing the sums in their head at the kitchen table.
Which is a strange thing to carry into a quarterly economic release. But there it is.
What you can actually do with this
Probably nothing about the number itself. That is being decided a long way above any one person, and pretending otherwise would be dishonest.
But there is a real difference between being squeezed and not knowing why. And the one lever most people still hold is a clear-eyed view of their own situation — what comes in, what goes out, what’s fixed and what isn’t. It’s unglamorous, and it’s also the thing that gets people through tight years. If it helps to see it laid out plainly, our free Budget Calculator will do it in a few minutes, with nothing to buy and nothing to sign up for.
Past that: keep your own record of what you’re worth in skill and reliability, because the market’s number for you and the truth about you are two different numbers. Say the tight parts out loud to somebody you trust, because carrying it silently is how a money problem quietly turns into a shame problem. And be careful with any explanation that requires you to be the failure. Sometimes the split just moved.
The economy will do whatever it does this quarter. You are still not a rounding error in it.
What do you think?
If a company posts record profits in the same quarter that its workers’ share hits a record low, is that a sign the business is being run well — or a sign something has gone out of balance? There’s a genuine case on both sides. Tell us how you see it in the comments.
Share this
- Workers’ share of the US economy just hit its lowest level since 1947 — and the economist who flagged it says AI isn’t the reason yet. That’s the part that stopped me: https://bgodinspired.com/index.php/social-issues-and-justice/labor-share-of-income-record-low/
- Turns out the “record low” for workers’ share of income has been reported four times this year. It isn’t a spike. It’s a floor that keeps giving way. https://bgodinspired.com/index.php/social-issues-and-justice/labor-share-of-income-record-low/
- The oldest written rules about work weren’t about growth or output. They were about paying the person before sundown, because they were counting on it that night. Still holds up: https://bgodinspired.com/index.php/social-issues-and-justice/labor-share-of-income-record-low/
Questions people are asking
What is the labor share of income?
The labor share of income is the portion of a country’s total economic output that goes to workers as wages, salaries and benefits, rather than to business owners as profit. It is normally written as a percentage of GDP. In the United States in the second quarter of 2026, it was 52.8% — the lowest level since this measure began in 1947.
Why is the US labor share of income at a record low in 2026?
Gregory Daco, chief economist at EY-Parthenon, attributes the decline mainly to productivity gains from automation and cost-cutting that are concentrated in large, capital-rich firms rather than spread across the wider economy. In a note to Fortune published on 4 September 2026, he also said he does not believe there is a floor beneath the current level, rejecting the assumption that the share would stop falling at 50%.
Is artificial intelligence causing the drop in workers’ share of income?
Not so far, according to the economist who flagged the record low. Gregory Daco of EY-Parthenon attributes the current gap to automation and cost-cutting rather than to artificial intelligence, and notes that the AI boom has not really begun. On that reading, AI is a possible future factor rather than the explanation for what has already happened.
How far has the labor share of income fallen since 1947?
Reporting on the long-run series puts labor’s share at around 70% when US records began in 1947, against 52.8% of GDP in the second quarter of 2026. That is a decline of roughly seventeen percentage points, though the precise gap depends on which measure of national output is being used. Over the same recent quarter, corporate profit margins reached a record 14.9% of GDP.
Does a record low labor share mean workers are being paid less?
Not necessarily in absolute terms. The labor share measures the proportion of total output reaching workers, not the size of any individual paycheck. Pay can rise while the share still falls, if the total value produced rises faster. What a falling share indicates is that a smaller slice of each additional unit of output is reaching the people doing the work.
