The job-hopping data
For a decade the fastest raise was the exit. Now that the switching premium has collapsed, the numbers say something quieter about what leaving a job actually buys.
For most of the last decade, the surest way to give yourself a raise was to quit. Not to work harder, not to ask, not to wait for the annual review where a manager apologizes for the budget. To leave. The person who walked out the door and signed somewhere new reliably out-earned the loyal colleague who stayed at the same desk doing the same good work. This was not a rumor traded at happy hour. It was one of the most stable patterns in the American labor market, measured every month, and for a long stretch it was enormous.
The clearest snapshot came in 2022, when inflation was eating everyone's paycheck at once. Pew Research found that among workers who changed employers between April 2021 and March 2022, the median switcher's real earnings rose 9.7 percent while the median worker who stayed lost 1.7 percent. Sixty percent of switchers came out ahead of inflation. Fewer than half of stayers did. Same year, same economy, same skills. The only variable that moved was whether you left.
The Atlanta Fed has tracked this gap since 1998 in its Wage Growth Tracker, which follows the same people over twelve months and sorts them into switchers and stayers. Through 2022 and 2023 the switcher line ran roughly two full percentage points above the stayer line, month after month. That gap has a plain-English name. It is the tax you pay for staying.
The premium was never a reward for disloyalty
It is tempting to read that number as proof that the market punishes loyalty and rewards restlessness, and to conclude that the winning move is simply to keep moving. That reading gets the mechanism backward. The raise you get by leaving is mostly a correction, not a windfall. It is the market repricing you to what you are currently worth, all at once, because a new employer has to bid at today's rate to get you and your old one never had to.
This is the quiet arithmetic of staying. Your pay is set on the day you are hired and then nudged upward in small annual increments that rarely keep pace with what your skills would fetch on the open market. Every year you stay, the distance between your salary and your replacement cost widens by a little. You do not feel it, because nothing is being taken from you. You are simply not being repriced. The switcher premium is not a bonus for jumping. It is the accumulated gap between what you were paid and what you were worth, handed back to you in a single lump the moment someone else has to compete for you.
The raise you get by leaving is mostly the money you were already owed, collected all at once because someone finally had to bid for you.
That reframing matters because it tells you when leaving works and when it does not. Switching pays when it corrects a real underpricing or a real mismatch between what you can do and what your job lets you do. It pays much less when it just relocates the same misalignment to a new building with a marginally bigger number attached. The premium was always doing two jobs at once, and only one of them was about money.
The lever just jammed
Here is what makes this the moment to think about it rather than a year ago. The great arbitrage of quitting has largely closed. ADP Research, which measures the same paychecks month over month for more than ten million workers, reported that by mid-2026 job-changers saw pay growth of 7.0 percent against 4.4 percent for job-stayers, a gap of about two and a half points, down from the high-single-digit premiums of the Great Resignation. The Atlanta Fed's tracker has at points shown the switcher advantage vanish outright, with stayers matching or beating movers for the first time since the years just after the 2008 financial crisis. The lever that reliably manufactured a raise for anyone willing to pull it is, for now, mostly stuck.
This is not a small shift in a footnote. It changes the whole logic of the mid-career decision. When leaving reliably paid double-digit gains, a lot of moves that fixed nothing about the work still looked smart on the paycheck. The premium papered over bad reasons for quitting and rewarded them anyway. With that cushion gone, a job change has to justify itself on its own terms again. You are no longer being paid a large fee to relocate your dissatisfaction. This is part of why so many people are choosing to sit still, a pattern we have called the great stay: the exit stopped being an obvious upgrade.
What switching actually buys
Strip out the boom-year premium and the deeper case for moving is not really about a single raise at all. It is about growth, and the growth shows up early. In a foundational study of young workers, economists Robert Topel and Michael Ward found that job changes account for at least a third of all wage growth in a worker's first ten years, a decade in which the typical person holds around seven different jobs. Early careers are built by movement. The moves are how you discover what you are good at, what you are worth, and which rooms you actually want to be in. Each change is a small experiment in fit, and the compounding of those experiments, not any one salary bump, is what bends a career upward.
By mid-career the physics change. The moves get rarer and the stakes get higher. American workers now stay with an employer a median of 3.9 years, the shortest tenure the Bureau of Labor Statistics has recorded since 2002, which sounds like a lot of churn until you realize it means the average job is barely long enough to master before the clock resets. The question stops being how often to move and becomes something harder: whether the move in front of you corrects a mismatch or merely trades a familiar one for a fresh one you cannot see yet.
The number that actually matters
So the job-hopping data ends up asking a different question than the one it seems to answer. It looks like a question about loyalty and money, about whether to stay or go. It is really a question about diagnosis. A raise from leaving is only durable when the new seat is a genuinely better match, when it uses more of what you are strong at and less of what drains you. The worker who feels overqualified and underused is not helped by a job that pays five percent more to underuse them somewhere else. The person weighing a bigger swing, a real change of direction, is not doing something reckless. A mid-career pivot is ordinary now, and the data on movement is on their side, as long as the movement is toward a better fit and not just away from a worse one.
That is the part the premium hid for a decade. When quitting paid double digits, you could leave for the wrong reason and still come out ahead. You cannot anymore. The collapse of the switcher premium is not bad news for anyone thinking clearly about their career. It is the market removing a distraction. The question was never whether to hop. It was whether the next thing fits better than the last one, and now, for the first time in years, that is the only question the paycheck is willing to answer.