Markets fixated on the June headline. Just 57,000 jobs added against a consensus of 110,000-114,000. It was ugly. But the bigger story sat one line lower in the Bureau of Labor Statistics release, and almost nobody caught it.

May’s payrolls, originally reported at 172,000, were quietly revised down to 129,000. That is a 43,000-job haircut, roughly 25% of the original print. The month that gave equity markets confidence and kept rate-hike bets alive was not the month we thought it was.

Put the two together and the picture sharpens considerably. The US labour market did not suddenly stumble in June. It was already weakening in May, and nobody knew it until this morning.

How BLS Revisions Actually Work

Every Non-Farm Payrolls release contains revisions to the prior two months. This is not a flaw in the data; it is the methodology working as designed. The first estimate captures roughly 60% of the establishment survey sample. The second revision brings that closer to 80%, and the third lands around 90%.

What this means in practice is straightforward: the initial NFP print is an educated guess. A good one, typically, but still a guess. Markets trade aggressively on that first number, and by the time the revision arrives 30 or 60 days later, nobody cares. Attention has moved on.

That is exactly what happened here. In early June, markets priced off the 172K headline. Rate-hike expectations firmed. Equity bulls pointed to resilient hiring. And then today, the BLS told us that resilient hiring was actually 25% weaker than advertised.

May Was Not the Month We Traded

When May’s 172K headline landed, the narrative was clear: the labour market was bending but not breaking. It was a Goldilocks print. Hot enough to suggest economic resilience, cool enough to keep the Fed from panicking about overheating.

But at 129K, the interpretation changes entirely. That is below the breakeven rate most economists estimate is needed to keep pace with labour force growth (roughly 150K-180K per month, depending on immigration assumptions). May was not Goldilocks. It was a miss, hidden by incomplete data.

And here is the uncomfortable question: if June’s 57K eventually gets revised downward as well, the two-month combined figure could land well below 150K. That is not a soft patch. That is a labour market that is decelerating faster than any single headline captured in real time.

The World Cup Factor: Seasonal Noise or Structural Signal?

Leisure and hospitality shed 61,000 jobs in June. On the surface, this looks alarming. It is one of the largest single-sector drags in recent memory. But context matters.

The FIFA World Cup, hosted across North American cities this summer, has pulled a significant portion of the hospitality workforce toward tournament-related employment that may not show up in establishment surveys the way traditional payrolls do. Workers relocating temporarily, picking up gig work, or shifting into event-related roles create gaps in the standard seasonal adjustment models the BLS applies.

This does not mean the -61K figure is fiction. It means a portion of it is likely seasonal distortion rather than a structural decline in hospitality demand. The distinction matters because it changes what the data tells us about the underlying economy.

Two Months of Weakness, Not One

This is the part most commentary will miss today. Traders will debate whether 57K is a one-off or the start of something worse. The revision answers that question before it is even asked.

May was weak. June was weaker. The trajectory is clear, even if the precise magnitude is muddied by World Cup effects. When you revise the prior month down by 25% and the current month misses by nearly 50%, you are not looking at a blip. You are looking at a trend that the initial data obscured.

Unemployment holding at 4.2% provides a partial offset. It suggests that while hiring has slowed, layoffs have not accelerated meaningfully. Companies are reducing their intake rather than cutting headcount. That distinction keeps this in “slowdown” territory rather than “recession” territory, at least for now.

But 4.2% is doing more work in this narrative than it should. The household survey (which drives the unemployment rate) and the establishment survey (which drives NFP) have been diverging for months. When they disagree, the establishment survey tends to be the better forward indicator. And the establishment survey just posted back-to-back soft prints.

What This Means for the Fed

Rate-hike probability for September dropped from 67% to 50% on this release. That repricing is rational but may not be aggressive enough if the revision pattern continues.

The FOMC Minutes land on 9 July. They will reflect the meeting that took place before this data was available, so do not expect direct commentary on today’s numbers. But the tone will matter. If the minutes reveal a committee that was already concerned about labour market softening, today’s data confirms those concerns retroactively. If the minutes show a committee still focused primarily on inflation, the market will have to do the repricing work itself.

Here is the key question for Fed watchers: how many soft prints does it take to shift the committee’s reaction function? One bad month can be dismissed. Two, especially when one was hiding in the revision, is harder to ignore. The dual-mandate framework requires the Fed to weigh maximum employment alongside price stability. When the labour side softens this quickly, the calculus changes.

What to Watch Next

• FOMC Minutes (9 July): The committee’s internal debate will signal whether labour market concerns were already on the table before this print arrived.
• Jobless claims: Weekly claims data will provide a real-time check on whether the NFP weakness is translating into actual separations or whether it remains a hiring slowdown story.
• July NFP revision of June: If June’s 57K gets revised lower, the two-month narrative intensifies. If it gets revised higher, the World Cup distortion theory gains credibility.
• Leisure and hospitality recovery: This sector should normalise after the World Cup concludes. If it does not, the structural concern deepens.