August 2026 Jobs Report Preview: Not Cracking, Still Frozen — and Now It's Costing Workers

Key Takeaways

• Frozen flows, not layoffs. The quits, hires, and layoff rates each sit about 17% below their 2018–19 averages. The weakness is showing up in hiring and mobility, not layoffs.
• Normal openings, fewer moves. Openings and the vacancy-to-unemployment ratio look near-normal. Far fewer workers move through them — about 15% fewer hires per opening.
• Security up, re-entry harder. Job security is high, but the unemployed have a harder time getting back in: about a 30% chance of finding work in a given month, down from 35% before the pandemic.
• The freeze reaches pay. Quits are at a cycle-low 1.9%. Wage growth has slowed to 3.2%, below 3.4% inflation; BLS estimates real average hourly earnings fell 0.2% over the year.

A frozen market, not a failing one

July’s JOLTS completes a two-year pattern. The quits (1.9%), hires (3.2%), and layoff (1.0%) rates have each fallen about 17% below their 2018–19 averages. This isn’t a market of mass layoffs. It’s one where almost nobody moves — hiring is low, quitting is low, and layoffs are low. Gross worker turnover — hires plus separations — is 6.4% of employment, well below the 7.6% pre-pandemic norm and back near the lows of the sluggish post-Great-Recession recovery.

Normal openings, frozen flows

Openings tell a calmer story than the flows do. The openings rate (4.4%) is close to its 2018–19 level, and vacancies per unemployed worker (1.05) sit just under the 1.16 norm — far below the 2.0 peak of 2022. But posted demand and actual hiring have pulled apart: there are now about 15% fewer hires per opening than before the pandemic.

The cost falls on outsiders

Incumbents remain relatively protected. The estimated monthly probability that an employed worker moves into unemployment is 1.4%, below the 1.7% pre-pandemic norm. There are about 1.8 quits for every layoff, close to the pre-pandemic ratio of 1.9. If you have a job, you are relatively secure. The burden lands on job-seekers.

The median unemployed worker has now been looking for almost 11 weeks, versus about 9 weeks in 2019. Outside the pandemic and its immediate aftermath, long-term unemployment is back around levels last seen roughly a decade ago. One in four unemployed workers has been out of work for at least 27 weeks, and an unemployed worker has only about a 30% chance of finding a job in any given month, down from 35% before the pandemic.

That lack of outside opportunity helps explain the low churn. Workers are less likely to quit when finding the next job is harder. And that comes with a cost: switchers have lost some of the outside offers that drive raises. In Topel and Ward’s classic study, job changes accounted for at least a third of early-career wage growth among the young men in their sample.

The freeze reaches wages

This is now visible in pay. Among labor-market tightness measures, the quits rate tracks wage growth best (Heise, Pearce, and Weber, 2026; reiterated by the New York Fed). With quits at a cycle low, average hourly earnings growth has slowed to 3.2% — below headline inflation of 3.4%. BLS estimates real average hourly earnings fell 0.2% over the year. Average hourly earnings are composition-sensitive, and fixed-weight measures are cleaner, but the direction is not in doubt. A stalled job ladder and softening pay are two signs of the same loss of worker bargaining power; the mechanism linking moves to raises is well understood (Moscarini and Postel-Vinay).

The wider costs: mobility and growth

Job switching is one of the ways workers move toward better matches and more productive firms. When outside opportunities disappear, fewer workers climb that job ladder — and productive firms have a harder time pulling workers away from less productive uses. Mortgage-rate lock adds another barrier: workers with cheap mortgages have more to give up if a better job requires a move.

That matters for long-run growth. Productivity doesn’t rise only because workers become more productive where they are. It also rises when workers move to firms where their skills, technology and capital can be used more effectively.

The labor market is another place where the AI payoff isn’t here yet. We’re spending enormous sums on technologies that could raise productivity, but hiring is weak, job switching is depressed, and workers are not being pulled toward new opportunities at anything resembling a boom-time pace. Installing the technology isn’t enough. Barring having the ability to work-from-home, workers also have to move toward the firms that use it most effectively. Some good news here: self-employment is on the rise.

The housing market feels the same freeze. Fewer job changes mean fewer moves and fewer home transactions, weakening an important source of housing demand. For builders, slower absorption is another reason to remain cautious about starting new homes. The freeze can also delay household formation: fewer opportunities to move up the job ladder mean weaker income growth and fewer reasons for young workers to relocate and set up households of their own.

We may be entering a period of extraordinary technological change with an unusually immobile labor market. An economy that gets worse at moving workers toward their most productive uses can also grow more slowly. That’s another reason to be skeptical of the Fed chair’s declaration that we’ve entered an era of “secular growth.” The investment boom is here. The labor-market payoff is not.

Bottom Line

The labor market is not cracking. It is frozen, and the freeze has stopped being costless. Layoffs are low and job security is high, but the flows that raise pay and move workers up have compressed, and real wages have turned negative. The jobs report will show whether low churn is starting to spill into outright employment weakness — falling payrolls, not just a stalled labor market.

For the Fed, the implication is important. There is little evidence that the labor market is generating inflationary pressure. Hiring is weak, quits are at a cycle low, and wage growth has slowed to 3.2%. With much of the remaining inflation pressure coming from supply-side shocks rather than excess labor demand, a September rate hike should not be the base case.

Numbers to Know — July 2026

Sources. BLS JOLTS and CPS via FRED (July 2026); BLS Real Earnings (July 2026). Constructed series (gross turnover, V/U, quits-per-layoff, job-finding and employment-exit probabilities) are author calculations. Davis & Haltiwanger (2014, NBER 20479); Heise, Pearce & Weber (2026, NY Fed); Moscarini & Postel-Vinay; Topel & Ward (1992, QJE).