Lucas Jackson / Reuters

The big-picture numbers that people rely on to describe how the economy is doing currently look pretty good: Unemployment is around 5 percent (which is considered very healthy by economists, because there will always be some people changing jobs in a good economy), and hundreds of thousands of jobs are being added every month.

But another important metric, wage growth, has been sluggish compared with pre-recession levels. On average, workers simply aren’t seeing their pay increase as quickly as it has at other times. According to data from the Bureau of Labor Statistics, real average hourly earnings increased by 1.8 percent in 2015. In a better year, earnings might rise 3 or 4 percent.

While a lot of workers are wondering when they’ll get their next raise, there are a number of Americans whose paychecks are increasing much more swiftly. In fact, if the BLS were to calculate earnings growth only for full-time workers who have been in the workforce for a while and aren’t retiring, it would come out closer to 4 percent—more than twice the average for all workers. “When you look at different sectors and different segments of the workforce, it really gives you a different picture of the wage growth,” says Ahu Yildirmaz, the lead economist at ADP Research Institute, the payroll processor that puts together the private-sector jobs report.

Among those full-time workers who are seeing healthy wage growth, those who switch jobs are doing particularly well. According to ADP’s data, full-time workers who changed jobs saw their paychecks increase an average of 4.5 percent, an improvement over the 3.9 percent average that covers all full-time workers. (In one extreme illustration of the power of switching jobs, the salary of one Alabama engineer increased by 31 percent over 4 years by changing jobs every six to 12 months.) This effect was even stronger in the midwest and northeast, where full-time-job switchers saw over 5 percent wage growth.

Breaking the data down by age, ADP also found that the wage increase from full-time-job switching was most pronounced for workers aged 25 to 34. “In general, younger job holders saw their wages rise much higher than the 35 [and above] workers,” says Yildirmaz, adding, “This is the time of life when workers acquire skills rapidly and enjoy frequent promotions. This dynamic slows considerably for mid-life workers, aged 35 to 54, and for 55 and above, it’s even slower.”

“We’re seeing that most of the sectors, the switchers are actually able to increase their wages,” says Yildirmaz. But the exception is the energy sector, in which full-time workers are taking lower-wage jobs in order to stay employed. Job switchers in construction and manufacturing also didn’t fare as well.

Job switching was not quite as promising for part-time workers, either. Generally, switching from a part-time job to another part-time job, or switching from part-time to full-time resulted in lower wage growth. Yildirmaz says that this indicates that part-timers are likely willing to accept lower hourly wages in exchange for more hours (and thus higher total income), and benefits.

The correlation of job switching and slightly higher wage growth is strongest for full-time young workers, and non-existent for part-time workers and those in struggling industries. What that suggests is the fact that it’s not the act of changing jobs that raises pay—it’s having skills that are valued in today’s economy and that allow workers to get and negotiate better job offers than the post-recession status quo.

Beyond how moving to a new employer can affect an individual worker, it can also send economists a message about the labor market more generally. Economists, aware that during economic expansion workers will switch jobs for a multitude of reasons—from getting higher pay to improving job satisfaction to seeking out more flexible schedules—have found that high rates of switching can foreshadow higher wage growth for other workers too.

This story is part of our Next America: Workforce project, which is supported by a grant from the Annie E. Casey Foundation.