Nonfarm payrolls rose by just 57,000 in June, less than half the 110,000 economists expected, and the U.S. Bureau of Labor Statistics stripped another 74,000 jobs combined from April and May on top of it. By the standard read on economic data, that is a soft report. The Dow Jones Industrial Average did not treat it that way. It added 395 points Thursday morning and touched a fresh intraday record, and the S&P 500 and Nasdaq each rose roughly 0.7% alongside it.
The unemployment rate ticked down, to 4.2% from 4.3%, but that decline arrived alongside a labour force participation rate that fell three tenths of a point to 61.5%, the kind of move that flatters a headline unemployment figure for the wrong reason: fewer people looking for work, not more people finding it. None of that slowed the rally.
The explanation is not that traders missed the details. It is that they were reading the report against a different reference point than the one a recession watcher would use.
The Reference Point Nobody Announced
Daniel Kahneman and Amos Tversky's research on framing, developed in their prospect theory work and extended in their 1981 paper on the framing of decisions, established that people do not evaluate outcomes in isolation. They evaluate them against a reference point, and the same objective number can register as a gain or a loss depending on where that reference point sits. A payrolls report is not exempt from this. Against a reference point of whether the economy is expanding at a healthy pace, 57,000 jobs and a shrinking labour force are a weak report. Against a reference point of whether the Federal Reserve will raise rates this month, the same report reads as relief: a softer labour market gives Chair Kevin Warsh less grounds to act on the hawkish rhetoric he has been building since taking office. Thursday's price action shows which reference point the market chose. The two year Treasury yield fell after the release, and traders who had been pricing meaningful odds of a July move pared them back.
Payrolls have swung between double digit misses and triple digit beats for twelve straight months, and June's release breaks a run of three consecutive beats with the sharpest miss since October.
Values shown are the headline print as first reported, not later-revised figures. April and May 2026 were each subsequently revised down; see body text.
Warsh Set the Anchor Three Weeks Ago
The reference point did not form on its own. Three weeks ago, at his first press conference as Federal Reserve chair, Warsh ended the practice of forward guidance and closed with a line markets read as a warning, that the committee would deliver price stability. The accompanying dot plot flipped hawkish, with nine of nineteen officials penciling in a hike by year end, up from none in March. Warsh repeated the message Wednesday from the ECB forum in Sintra, Portugal, sharing a panel with Bank of Canada Governor Tiff Macklem: prices, he said, are too high. Once that anchor is in place, a weak jobs report does not have to be good news on its own merits. It only has to be less alarming than the anchor implied. Canadian portfolios inherit this same read whether or not a Canadian number moved. The TSX reopens this morning after sitting out Wednesday's Canada Day holiday while Wall Street set the tone twice, once lower on Warsh's remarks, then sharply higher on the jobs miss. The index returns to trading having missed the framing shift in real time and must price both moves in a single session.
What the Rally Is Not Pricing
The details of the June report support the labour market framing more than the Fed relief one. Leisure and hospitality lost 61,000 jobs, a reversal that erases most of the sector's gains from earlier in the year. The prior three months, which had each beaten consensus and built a narrative of resilience, are now smaller than first reported by a combined 74,000 positions. Private payrolls from ADP, released a day earlier, showed just 98,000 new jobs against expectations near 120,000. None of this contradicts the market's rally. It simply was not what the rally was measuring. Reference point framing does not require the underlying data to be good. It only requires the data to be less threatening than whatever anchor the market had already set. That gap, between what a number contains and what a market chooses to measure it against, is where mispricing risk concentrates, and it closes only when a subsequent report forces a new reference point onto the table.