Job switchers versus job stayers: what the Atlanta Fed data shows about pay growth

If you're weighing whether to ask for a raise or start looking elsewhere, the Atlanta Fed's Wage Growth Tracker gives you a real benchmark: how fast pay is growing for people who stay put versus people who change jobs. Here's what the latest numbers say, how the gap has swung over the years, and how to use it in your next pay conversation without overselling what it proves.

What the Wage Growth Tracker measures

The Federal Reserve Bank of Atlanta builds the tracker from the Current Population Survey (CPS), the monthly household survey the Census Bureau runs for the Bureau of Labor Statistics. Households stay in the survey for four months, drop out for eight, then come back for four more. People are asked about their earnings in the fourth and the final month, which means the same person reports their pay twice, exactly 12 months apart.

The Atlanta Fed matches those two answers, works out each person's percentage change in hourly pay, and then takes the median of all those individual changes. The median is simply the middle value: line everyone up from smallest raise to biggest, and the person in the middle is the median. Half the group did better, half did worse.

That detail matters more than it sounds. The tracker is not "how much the typical wage went up." It is "how much the typical person's own wage went up." The Atlanta Fed spells out that these are different things: growth in the median wage compares whoever sits in the middle of the pay scale this year with whoever sat there last year, and those are almost certainly different people.

A few more things to know about how it's built:

• It's hourly pay. Weekly earnings are converted to an hourly rate using usual hours worked. Pay is before taxes and deductions, and it covers your main job only.
• It's nominal. There's no adjustment for inflation, so a raise that merely keeps pace with prices still shows up in full.
• It's smoothed. The headline is a three-month moving average, meaning each month's figure averages that month's median with the two before it. The breakdowns by age, industry and wage level in the downloadable spreadsheet are 12-month moving averages, which are steadier but slower to turn.
• The sample is modest. The matching produces about 2,000 individual wage-change observations a month, according to the Atlanta Fed's methodology notes.

Who counts as a "job switcher"

The tracker labels you a switcher if, compared with a year ago, you're in a different occupation or industry, or if you changed employers or job duties in the past three months. Everyone else is a stayer.

Two quirks follow from that. First, someone who got a new title and duties at the same company can land in the switcher group. Second, because the CPS surveys addresses, people who move house drop out of the data, so switching here only captures local moves. A worker who relocated across the country for a better offer isn't in the numbers at all.

Who is left out

The Atlanta Fed excludes people whose earnings are top-coded, which since 2003 means pay above an annualized $150,000. It also drops self-employed people (the CPS doesn't collect their earnings), anyone whose earnings were imputed, agricultural workers, and hourly pay below $2.13, the federal tipped minimum wage. The matched group also skews a little older and more educated than the workforce as a whole, because you need earnings in both years to be included.

The latest gap between switchers and stayers

The most recent release, updated October 8, 2026, covers September 2026. Here are the headline three-month figures:

So in September 2026, the median switcher's hourly pay was 4.7% higher than a year earlier, against 3.4% for the median stayer. That's a gap of 1.3 percentage points. In August 2026 it was 1.4 points.

Both groups slowed from August to September, but the switcher premium held up. If you only remember one number, remember that gap, not the headline 3.9%.

The 12-month moving averages for September 2026 in the Atlanta Fed's spreadsheet tell a similar but softer story: 3.5% for stayers and 4.4% for switchers. The longer average is still catching up with a premium that reopened over the past year.

For context from a different angle, the BLS Employment Cost Index, which tracks what employers pay for a fixed mix of jobs rather than following individual people, showed private industry wages and salaries up 3.1% in the 12 months ending June 2026. After inflation, those private wages were down 0.4% over the year. The September 2026 ECI was scheduled for October 30, 2026, so check the BLS release for the newer figure.

How the gap has changed over time

The switcher premium is not a fixed bonus. It stretches and shrinks with the job market, and it has even flipped below zero.

Here are snapshots from the Atlanta Fed's own monthly releases, all three-month moving averages, labelled with the month the data covers:

The long view

The longer series in the Atlanta Fed spreadsheet, using 12-month averages, shows the pattern clearly:

• Recession years: In early 2010, switchers were actually behind. In February 2010 the 12-month average was 2.0% for switchers against 2.6% for stayers. When jobs are scarce, people who move often do so because they have to, not because someone outbid their employer.
• The 2022 hiring boom: The overall tracker hit a high of 6.7% in June and July 2022, according to an Atlanta Fed analysis. The switcher premium peaked a few months later: by November 2022, the 12-month average was 7.7% for switchers against 5.5% for stayers, a 2.2-point gap, the widest in the series going back to 1997.
• The cool-down: By September 2025, the 12-month figures had nearly converged at 4.3% for switchers and 4.1% for stayers. In the three-month figures, stayers were briefly ahead in early 2025.
• The reopening, with a wobble: The gap widened again from late 2025 into early 2026, then nearly closed in spring. In the current spreadsheet, May 2026 shows 3.7% for switchers against 3.3% for stayers on the three-month measure. By August and September 2026 it had opened back up past a full point.

The takeaway for timing: the premium for moving tends to be biggest when employers are competing hard for workers, and it can vanish when hiring cools. BLS turnover data is a useful companion signal. In August 2026, quits held at 3.1 million, a rate of 1.9%. Quits are mostly voluntary, so BLS treats that rate as a gauge of how willing or able people are to leave.

Interesting wrinkle: right now the gap is fairly wide while overall pay growth is slowing. That suggests employers are still paying up to fill roles they truly need, while being stingier with raises for the people they already have.

Pay growth by age, wage level and industry

The Atlanta Fed also breaks the tracker out by worker group. These come from the downloadable spreadsheet and are 12-month moving averages, so they move slowly. Here's September 2026 next to September 2025:

A "wage quartile" just means a quarter of the workforce, ranked by pay. The lowest quartile is the bottom 25% of earners, the highest is the top 25% (minus anyone above the $150,000 top-code).

What jumps out:

• Young workers always run hottest. People early in their careers get bigger percentage jumps because they're learning fast, changing roles often, and starting from lower pay. A 16-to-24 figure well above the overall rate is normal, not a sign that everyone should expect it.
• Older workers run coolest. The 55+ group consistently sits below the overall tracker. If you're in that bracket, comparing yourself to the headline number can make a normal raise look small.
• Low earners lost their edge. In September 2022, the bottom quartile was growing at 7.6% on the 12-month measure, against 4.4% for the top quartile. By September 2026 the lowest quartile was slightly behind the others. The tight market that pushed up low-wage pay has eased.
• Industry matters a lot. In September 2026, the 12-month figure for construction and mining was well above education and health or leisure and hospitality. Your sector's line is a better comparison than the national headline.

Keep in mind each group is only a slice of the roughly 2,000 matched observations a month. The Atlanta Fed has noted that small slices such as wage quartiles are noisy, which is why it shows them as 12-month averages. Treat moves of a few tenths of a point as background static.

What the data can't tell you about your own situation

This is the honest part. A median tells you about the middle of a crowd. It says nothing certain about where you'll land.

The spread around that middle is huge. The Atlanta Fed notes that the 75th percentile of individual wage growth (the point where 75% of people did worse and 25% did better) typically sits about 10 percentage points above the median, while the 25th percentile sits about 5 points below. In 2022, an Atlanta Fed analysis found that around a quarter of people in the sample had raises above 18%, while another quarter had zero or negative growth. Same economy, same year, wildly different outcomes. Even in calmer times, plenty of people get nothing: in September 2026, 13.4% of the matched sample had a pay change within half a percent of zero, per the Atlanta Fed spreadsheet.

Here's what else the tracker can't see:

• Why people switched. It doesn't separate people who jumped for a raise from people who were laid off and took what they could get. Both are "switchers."
• Total pay. Bonuses, equity, 401(k) matches, health premiums and remote-work flexibility aren't in an hourly wage figure. A switch that adds salary but costs you a richer retirement match or cheaper health coverage may not be a win.
• High earners. Anyone above the $150,000 annualized top-code is excluded, so if you're in that band, this isn't your benchmark.
• Long-distance moves. People who relocated for a job drop out of the sample.
• Your employer's budget, your performance, your market. None of that is in a national median.

Also note a data hiccup: the Atlanta Fed says its sample was about 10% smaller in November and December 2025 than in September 2025, likely because of falling CPS response rates, and October 2025 is missing entirely because of the federal government shutdown. Figures from that stretch carry wider error bands. Past months can also be revised a little: the January 2026 switcher figure was first published as 4.7% and shows as 4.8% in the current spreadsheet. When you quote a figure, quote it from the latest release and name the month.

So use the switcher-stayer gap as a sense of the market's weather, not a forecast for you. It tells you whether the wind is at your back when you move. It doesn't tell you whether a particular offer will be good.

Using it in a raise or offer conversation

The tracker is most useful as a neutral, credible reference point. It's published by a Federal Reserve bank, it's free, and it's updated monthly, usually by the second Friday. That makes it a calmer opener than "I saw on social media that everyone's getting 10%."

Worked example (hypothetical)

Say you earn $60,000 and your manager offers a 3% raise. Using the September 2026 three-month figures:

The median stayer's raise is $240 more than your offer, and the median switcher's is $1,020 more. That's not a guarantee you'd get either amount, but it tells you a 3% offer sits below the middle of what people staying in their jobs saw over the past year.

Script: asking for a raise at review time

"I've been looking at the Atlanta Fed's Wage Growth Tracker. For September 2026, the median worker who stayed in the same job saw about 3.4% hourly pay growth over the year. Given what I've delivered this year, [one or two concrete results], I'd like us to look at something at or above that, closer to [your number]."

Why it works: it anchors to a neutral source, ties the ask to your results rather than the economy alone, and names a number.

Script: the stay-or-go conversation

If you have an outside offer, or you're seriously considering one, the switcher-stayer gap gives you a way to frame the conversation without sounding like a threat:

"I want to be upfront: I've been approached about another role, and I'd rather stay. The Atlanta Fed's data shows people changing jobs are seeing noticeably faster pay growth than people staying, about 4.7% versus 3.4% in the latest release. I'm not asking you to match the market blindly. I'm asking whether there's room to close part of that gap so staying makes sense for both of us."

Only use this if the outside interest is real. Bluffing with a data point won't hold up if they call it. For more on handling the employer's response, see our negotiation coverage, including guidance on raises and counteroffers.

Pair the national number with local data

The tracker is national and broad. Before you walk in, also look at pay ranges posted for similar roles near you, especially in states that require pay ranges in job ads. A specific range for your job title in your city carries more weight than a national median. Our salaries by role and salaries by city pages are a good place to start, and pay trends tracks the bigger picture month by month.

What to do next

• Check the newest release. The tracker updates monthly. Look up the latest stayer and switcher figures on the Atlanta Fed Wage Growth Tracker page the week of your conversation, and quote the month they cover.
• Find your slice. Use the interactive chart or spreadsheet to see the line for your age group, industry or wage level. It's a fairer comparison than the headline.
• Run your own math. Multiply your salary by the stayer and switcher rates to see the dollar gap. Then add the things the tracker ignores: bonus, match, benefits.
• Watch the trend, not one month. A gap that has widened for several months in a row says more than a single reading.
• Know when it's not your benchmark. If you earn above $150,000, are self-employed, or are relocating, look for data that covers your situation instead.

The numbers can tell you whether the market is rewarding movers right now. Whether moving is right for you is still a call only you can make.