Job Switchers Just Pulled Wage Growth Back Above 4 Percent

The Atlanta Fed's Wage Growth Tracker put median wage growth at 4.1 percent in August, up from 3.8 percent in July, according to data published this week by the Federal Reserve Bank of Atlanta. That single tick matters more than it looks. It is the first move back above 4 percent since spring, and it is not coming from across-the-board raises. It is coming almost entirely from people who changed jobs.

The Job Switcher series tells the story. Median wage growth for workers who switched employers or roles hit 5.0 percent in August, a jump of six tenths of a percentage point in a single month and the highest reading since January. Workers who stayed in their jobs saw 4.3 percent, up just two tenths. A year ago, in August 2025, the overall tracker stood at 4.0 percent with switchers at 4.5 percent, so the switcher premium has widened even as headline growth sits roughly where it did twelve months back. Go back two years, to August 2024, and the tracker read 4.7 percent overall. The level is lower now, but the August direction is the opposite of what a softening labor market would normally deliver.

That combination is odd enough to be worth sitting with. Weekly initial claims have held between 206,000 and 212,000 since early August, with the week ending September 5 at 206,000 per Labor Department data, and JOLTS hiring recently touched a six-month low. Firms are not adding headcount at the pace of 2022 or 2023. Yet the workers who do move are still extracting meaningful raises, and the spread between switchers at 5.0 percent and stayers at 4.3 percent says employers would rather pay up for someone proven than lift pay across the whole payroll.

The distribution underneath the average carries the second finding. The tracker's lower half of the wage distribution, workers paid below the median, grew wages at 5.0 percent in August, against 4.0 percent for the upper half. Wage compression is still running, which is unusual this late in a cycle. Low-wage workers typically catch up during tight labor markets and lose ground quickly once hiring cools. August says the catch-up phase has not fully ended, at least for the people who manage to switch jobs. For workers who stay put, the math against prices is less friendly. The Bureau of Labor Statistics put average hourly earnings at $37.75 in August, up 3.1 percent year over year per FRED's CES0500000003 series, against an August CPI print of 3.4 percent. Stayers on median pay are roughly running in place.

What does this do to the Fed picture. A 4.1 percent median wage print, rising rather than falling, is the kind of number that keeps the inflation-hawk half of the Federal Open Market Committee talking. The committee has leaned all year on the argument that cooling wage growth gives it room to ease. The August tracker does not break that argument, because the twelve-month average has drifted from 4.2 percent at the start of the year to 3.8 percent now. But it complicates the October narrative. If September's tracker also prints above 4 percent, the case that labor income pressure has fully normalized weakens, and the bar for the next cut rises. Real Wages Fall Behind 3.4% Inflation as August CPI Outpaces.

Sector texture supports the read. The tracker's college-degree series moved from 3.9 percent in July to 4.1 percent in August, and the 25 to 54 prime-age series rose from 4.0 to 4.4 percent, per the Atlanta Fed file. Pay growth accelerated across education and age cuts rather than in one corner of the market, which argues against dismissing the move as a sampling quirk. The one soft spot is services, where growth ticked up only two tenths, less than the overall gain.

The counterargument deserves its say. One month of a survey-based, three-month moving average is not a trend, and the tracker is built from the Current Population Survey with a rotating sample, so single-month jumps partly reflect who happened to be asked. The job-switcher series in particular is the noisiest line on the page. It is possible September's print retraces most of the August jump and the twelve-month average keeps gliding down. That is genuinely the base case if hiring keeps cooling, because fewer job changes eventually means fewer people collecting the switcher premium.

For households, the practical takeaway is about timing rather than averages. The gap between the 5.0 percent switcher series and the 4.3 percent stayer series is the widest since last winter, per the Atlanta Fed data, and it is a measurable cost to staying put. Workers who changed employers this year captured roughly seven tenths of a percentage point more annual pay growth than those who did not, on the tracker's medians. That premium historically compresses when hiring freezes spread from headcount to backfills. Until claims break above 250,000 on a sustained basis, the data do not show that happening. See Labor Flows Point to Slack Ahead of September Payrolls Print.

The calendar from here is short. The next weekly claims release lands Thursday morning at 8:30 a.m. Eastern, covering the week ending September 12. The September Employment Situation arrives Friday, October 2, and the September tracker follows in early October with its own read on whether August was noise or signal. Between now and then, watch the quit rate in the JOLTS release, because if quits keep sliding while switcher wage growth holds above 5 percent, the pay premium is concentrating among ever fewer movers, which is the shape a labor market takes just before it stops being tight at all.

The implication is straightforward for anyone watching either wages or policy. The headline labor market is cooling, and the claims data show no outbreak of layoffs. But pricing power has not left the market. It has narrowed to the people changing seats, and in August they got paid like it. August Job Cuts Hit 52,881, Lowest August in Four Years.