Switching jobs to get a real raise is advice old enough to be a cliché, and the latest numbers say it still holds up. The Federal Reserve Bank of Atlanta's Wage Growth Tracker showed median pay growth of 5.0 percent for job switchers in August 2026, compared with 3.6 percent for people who stayed put. Payroll data from ADP Research points the same direction. In its September report, base pay for job-changers rose 4.8 percent year over year, while base pay for job-stayers rose 3.0 percent. Counting bonuses, commissions, tips and other earnings, the gap widens to 7.3 percent versus 4.4 percent.
That is the conventional read, and it is correct as far as it goes. What it does not explain is why the gap exists. Companies are not stingy by temperament, and they are not blind to the market. The gap exists because two different clocks set the two numbers. Your raise is priced once a year, months in advance, against a budget. A new employer's offer is priced today, against whatever the market is paying this week. When those two clocks drift apart, the difference shows up as the switching premium.
The raise was priced months ago
Start with how raises get funded. Marsh's Mercer surveyed 1,001 U.S. employers in July 2026 and found they plan to set merit increases at an average of 3.2 percent for 2027, with total salary increases averaging 3.5 percent once promotions, cost-of-living adjustments and other changes are included. Mercer's separate planning analysis puts actual total increases for 2026 at 3.4 percent. Marsh noted that if projections hold, 2027 will be the fourth consecutive year of moderate increases. Those figures are averages across employers, including some that report no increase at all, so they describe a typical budget rather than a promise to any one person.
Notice the timing. As of July, 87 percent of those organizations said their 2027 budgets were still preliminary, 8 percent had proposed a figure to leadership, and only 5 percent had secured final approval. In other words, the number that will govern your raise next year was being drafted before anyone knew what hiring conditions would look like in the spring. Budgets are built on forecasts and last year's actuals, so they respond to the market slowly.
Industry matters too. Mercer found the largest merit budgets in high tech at 3.8 percent and banking at 3.7 percent, while consumer goods came in at 2.9 percent and healthcare and retail at 3.0 percent. If you work in one of the lower-budget sectors, the ceiling on a stay-put raise is lower before your manager says a word.
The offer is priced today
Now look at the other side of the table. Most surveyed employers manage pay through a formal salary structure, a set of ranges tied to each job level. Mercer found that 75 percent do this and adjust the structure annually, and that the adjustment is typically a bit smaller than the merit budget. So the range attached to your role tends to move by something close to three percent a year.
The market price of that role does not have to move in step. If demand for your skills rises faster than the structure does, the range lags behind. A recruiter filling an open seat is not working from last year's structure. A recruiter is competing against other offers on the candidate's desk right now and has to price accordingly. That is how a company can sincerely believe it pays fairly and still lose someone to a competitor offering a much larger jump, because both of those things can be true at once.
ADP's chief economist, Nela Richardson, said when the July figures came out that job-changers are "highly sensitive to real-time economic conditions." She tied their rapid pay growth to supply constraints in parts of the labor market. Stayers, by contrast, are paid from a budget set well before those conditions arrived. That difference in timing is the core of the mechanism.
The premium comes and goes
If the mechanism were permanent, the premium would be permanent. It is not. The Atlanta Fed first reported job-stayers at 4.1 percent and job-switchers at 4.0 percent for July 2025, which means that for a stretch staying put paid slightly better than switching. Its next release showed switchers back ahead, 4.4 percent to 3.8 percent. The Fed's figures are subject to revision, but the pattern is clear. ADP Research reported something similar in January 2026, when it said the premium for switching in its data, which goes back to 2020, was the smallest it had recorded. By late summer the gap had widened again.
The premium rises and falls with how tight the labor market is. When employers are scrambling to fill roles, offers jump and the gap opens. When hiring cools, offers flatten and the gap narrows. ADP also says that pay growth overall has been decelerating for the past four years, which is part of why every raise, stay or switch, looks smaller than it did a few years ago.
Where the headline overstates it
Before you update your résumé, a few caveats are worth keeping in mind.
First, definitions matter. The Atlanta Fed counts someone as a job switcher if they are in a different industry or occupation than a year earlier, or if their job duties or employer changed in the last three months. That is broader than leaving for another company, so part of the premium may come from people who changed roles. The tracker is also built from the Current Population Survey, using the median of individual wage changes, with about 2,000 matched observations a month. The Fed notes that its sample was about 10 percent smaller in November and December 2025 than in September 2025, likely because fewer households responded to the survey. The numbers describe the typical mover, not what you will be offered.
Second, the market is cooling. The Bureau of Labor Statistics reported that job openings slipped to about 7.1 million in August, a change it described as small, and that the quits rate held at 1.9 percent, well below the 3.0 percent peak recorded in November 2021. Fewer people are leaving jobs than during the post-pandemic scramble. Those who do move are not a random sample of the workforce, so the average gain they report is not a guarantee for everyone who tries.
Third, not every employer pays new hires more. Ravio, a pay-benchmarking firm, analyzed mid-level software engineers at European companies on its platform and found that from January 2024 through March 2026, new hires out-earned existing employees by only 0.5 percent on average. New-hire premiums spiked in February, at 14 percent in 2025 and 12.3 percent in 2026, and existing employees pulled ahead for several months after mid-year pay reviews took effect. The data is European and limited to one job family, so it is a caution rather than a rule, but it shows that the gap between new and existing pay can close.
Fourth, a higher base is not the whole picture. ADP notes that base pay changes tend to be long-lasting, while its gross pay measure folds in bonuses, commissions, tips and other earnings that can vary. Unvested equity, benefits that grow with tenure and the standing you have built with your current employer may all stay behind when you go. Weigh the full package, not just the headline number.
How to capture the gap without leaving
The more useful finding is that employers have built-in ways to close the gap, and many of them use them. In the Mercer survey, 64 percent of organizations said they have provided or will provide off-cycle salary adjustments in 2027. Mercer's own reading is that promotions and off-cycle pay "play an important role in total rewards and career progression." Promotions are a narrower door than they sound, though. Employers expect to promote about 8.4 percent of their workforce in 2027, down from 8.6 percent in 2026 and 9.9 percent in 2025.
Timing matters here. Mercer's planning analysis notes that budgets are closer to final by its October survey than they were in July, which suggests the window for shaping them is narrowing. Coming with current market data for your role, rather than a general sense that you deserve more, gives a manager something concrete to bring to compensation. A change to your base pay or level would also carry forward, which a one-time bonus does not.
If the answer is no, the logic of the two clocks tells you how to read it. A company that cannot or will not move your pay toward the market rate is telling you something about how fast its internal clock runs relative to the outside one. At that point the decision to switch is a response to a price signal rather than an act of disloyalty.
The bottom line
The headline that people who switch jobs get bigger raises is true, and the numbers behind it are solid: 5.0 percent versus 3.6 percent at the Atlanta Fed in August, 4.8 percent versus 3.0 percent in ADP's September base pay data. But the interesting story is not a lesson about loyalty. It is a lesson about timing. Your employer sets your raise from a budget written long before you asked, while the market sets a new hire's pay from conditions that exist today. Wherever those two numbers diverge, the premium appears, and wherever they converge, it shrinks.
Knowing that changes how you approach the decision. You can treat the annual review as the floor rather than the ceiling, ask for a market adjustment while budgets are still being shaped, and treat an outside offer as information about what your skills are worth right now. And if your employer will not respond to that information, you will at least know that the premium is not a mystery. It is the price of a clock that runs slow.
