Trabajos blandos, salarios altos: El informe de diciembre que preocupa a los mercados de bonos

The December jobs report landed with the kind of “mixed, but market-moving” profile that can keep both doves and hawks busy on TV for the rest of the day. Headline nonfarm payrolls rose by 50,000, below expectations (most estimates clustered around ~70–73k) and below the downwardly revised November gain of 56,000. At the same time, the unemployment rate ticked down to 4.4% versus 4.5% expected, while wage growth ran hotter than forecast on a year-over-year basis (average hourly earnings up 3.8% vs. 3.6% expected). The result is a report that reads soft on hiring momentum, firm on labor-market tightness at the margin, and sticky on the “last mile” inflation narrative through wages.

Rates told the story in real time. The initial reaction in the 10-year yield was a brief pop to 4.211%—notable as the highest level since September 4—before the broader market settled into a more two-handed interpretation: slower job creation argues for easier policy later, but hotter wages and a still-resilient unemployment rate make it harder to price an imminent easing cycle. In other words, the bond market saw the headline payroll miss, then immediately remembered wages exist. Equity futures leaned positive after the release, but the rates tape kept a cautious tone, consistent with a market that’s increasingly allergic to “higher wages, fewer cuts.”

Starting with the establishment survey, payroll growth in December was modest and concentrated in the usual late-cycle stalwarts. Food services and drinking places added 27,000 jobs, health care added 21,000 (including a 16,000 gain in hospitals), and social assistance rose 17,000 (mostly in individual and family services). Retail trade, meanwhile, lost 25,000 jobs, with weakness concentrated in warehouse clubs/supercenters/general merchandise (-19,000) and food and beverage retailers (-9,000), partially offset by gains in electronics and appliance retailers (+5,000). Elsewhere, most major categories showed little change, including construction, manufacturing, professional and business services, financial activities, transportation/warehousing, and mining/energy extraction.

The revisions mattered, too, and they leaned in the wrong direction for momentum. October payrolls were revised down to -173,000 from -105,000, while November was revised down to +56,000 from +64,000. Combined, October and November were 76,000 lower than previously reported. That’s not catastrophic, but it reinforces the idea that the labor market ended 2025 with a weaker hiring pulse than the early reads suggested. Zooming out, the BLS noted payroll employment rose by 584,000 in 2025—an average monthly gain of 49,000—far slower than 2024’s 168,000 average. Put simply: job growth didn’t fall off a cliff, but it definitely took the elevator down a few floors.

Wages and hours were the “sticky” side of the ledger. Average hourly earnings rose 0.3% m/m, right in line with expectations, but the year-over-year pace accelerated to 3.8% (above the 3.6% forecast and above the prior pace). That’s the kind of detail that makes it harder for markets to fully embrace a near-term cut story, especially with inflation still a live topic and services still sensitive to labor costs. Meanwhile, the average workweek edged down to 34.2 hours from 34.3 (manufacturing fell to 39.9 hours), which is consistent with slower labor demand without outright layoffs—classic “no-hire, no-fire” behavior.

The household survey added a second layer of nuance. The unemployment rate fell to 4.4%, while the labor force participation rate edged down to 62.4%. The employment-population ratio ticked up to 59.7%. Underemployment improved: U-6 fell to 8.4% from 8.7%. But beneath the surface, longer-duration stress continues to build. The share of unemployed who have been jobless for 27 weeks or more rose to 26.0%, and the level of long-term unemployed is up 397,000 over the year. Part-time for economic reasons held at 5.3 million but is up 980,000 over the year, and the number of people not in the labor force who want a job rose 684,000 over the year to 6.2 million. This is the “softening without breaking” pattern: fewer hires, limited fires, but a growing pool of people finding it harder to get reabsorbed.

Where this report really hit markets is in the policy path it implied. Expectations for a Fed cut in April have slipped to 52%, and that matters because, historically, the market typically doesn’t treat a cut as “real” until probabilities are comfortably above 60%. You flagged the key point: we’ve never really seen the Fed move with sub-60% odds priced—meaning the market is now effectively signaling “April is no longer the base case.” The first cut is increasingly priced for June, with the curve still implying two cuts for the year and the second one clustering around September. That’s a meaningful shift in sequencing: the market isn’t scrapping easing altogether, but it is pushing it out, and it’s doing so in a way that keeps longer-end yields sensitive to every incremental inflation/wage input.

Putting it together, December’s report doesn’t scream recession, but it does validate cooling momentum. Payrolls undershot, revisions were negative, hours slipped, and hiring looks narrowly concentrated. Yet the unemployment rate improved, underemployment eased, and wages ran hot enough to keep the Fed cautious. The cleanest takeaway is that this is a late-cycle labor market: slower job creation, more friction for job seekers, and still-firm compensation trends. And for markets, that’s exactly the combination that can keep yields elevated even when the headline number disappoints—because the Fed doesn’t cut on “soft,” it cuts on “soft enough.”