In an average year, 5.9 percent of American workers move to a different occupation, and another 4.7 percent leave the labor force. That is the all-occupation baseline from the Bureau of Labor Statistics (BLS) 2024–34 Employment Projections, the same dataset agencies use to forecast openings. One worker in seventeen changes what they do. The interesting part is how unevenly that churn is distributed.
We joined the per-occupation rates to the 153 occupations our instrument tracks. The spread runs from pharmacists at 1.3 percent a year to tutors at 8.1. Same economy, six-fold difference in the odds that a given colleague is gone next year.
The stickiest careers in our set
| Occupation | Transfer /yr | Exit /yr | Posted median |
|---|---|---|---|
| Pharmacist | 1.3% | 2.4% | $99,140 |
| Lawyer | 1.4% | 1.8% | $137,576 |
| Physical therapist | 1.5% | 2.1% | $99,607 |
| Nurse practitioner | 2.1% | 2.2% | $86,521 |
| Registered nurse | 2.1% | 2.8% | $73,616 |
| Psychologist | 2.2% | 2.7% | $119,728 |
| Actuary | 2.6% | 1.8% | $122,291 |
Every one of those is licensed, and not casually licensed: doctorates, bar exams, actuarial exam sequences that take most of a decade. Then look three rows further down our ranking and the story complicates. Software engineers transfer at 3.2 percent a year, stickier than registered nurses, with no license at all. Whatever is holding people, it is not only the credential.
The revolving doors
| Occupation | Transfer /yr | Exit /yr |
|---|---|---|
| Tutor | 8.1% | 9.0% |
| Chef | 7.8% | 3.4% |
| Customer support specialist | 7.4% | 5.6% |
| Medical assistant | 7.4% | 4.5% |
| Pilot | 7.2% | 3.8% |
| Paralegal | 7.2% | 3.2% |
| Flight attendant | 7.1% | 6.4% |
Tutors turn over almost completely: 8.1 percent transfer out and another 9 percent exit the labor force every year, which is what you would expect from a role that is often a between-things job rather than a destination. Chefs and medical assistants churn despite holding certificates. The pilot figure looks strange until you notice the BLS category mixes commercial pilots (crop dusting, charters, instruction) with the airline majors; the instability lives at the commercial end.
Two different ways to leave
The table separates transfers (moved to a different occupation) from exits (left the labor force), and the two tell different stories about the same job. Tutors post the highest exit rate in our set, 9 percent a year, because tutoring is frequently a role people pass through on the way to school, parenting, or retirement. Flight attendants exit at 6.4 percent, nearly as high, an occupation people age out of or step away from rather than convert into something adjacent. Chefs are the reverse case: only 3.4 percent exit, but 7.8 percent transfer, meaning chefs who leave kitchens overwhelmingly keep working, just not as chefs. High exit reads as attrition from work itself; high transfer reads as skills going somewhere else. If you are planning a pivot, the second population is your competition and your proof of feasibility at once.
Separations are also where job openings actually come from. The same BLS table projects 18.9 million openings a year across the economy, and the overwhelming majority are replacement needs, someone transferred or exited, rather than newly created positions. Every churny occupation on the second table is, from the other side of the desk, an occupation that is constantly hiring.
Because here is the twist worth keeping: licensing by itself barely matters. Licensed occupations in our set average 4.2 percent annual transfer; unlicensed, 4.4. What separates the sticky from the churny is what the position cost to reach and what it pays to stay. A pharmacy doctorate is a moat. A food handler’s certificate is a formality. Both are licenses.
Medical assistant is the list’s most instructive entry, because its exits are not random. The observed-flow data shows departing medical assistants overwhelmingly become registered nurses, the ladder working as designed. What they almost never do is jump straight to nurse practitioner, despite a 76 percent skill match. One rung at a time, license by license. We wrote up that full pattern in “Where people actually go.”
Where this leaves you
Your occupation’s transfer rate is your base odds, the prior before anything about you. An architect’s 3.3 percent means leaving is unusual, which cuts both ways: fewer competitors on the way out, and less well-worn paths (the strongest observed one leads to interior design). A paralegal’s 7.2 percent means the door is already revolving, and the question is not whether people leave but where they land best. The instrument now carries the rate for every origin we track. Check your base rate before you decide you are stuck, because in several of these occupations, statistically, nobody is. Each occupation’s full pay distribution, and the trend behind it, now lives on its salary page.
Transfer and exit rates: BLS Employment Projections, Table 1.10, 2024–34 vintage, annual averages, 832 detailed occupations, public domain, joined to 147 of our 153 tracked occupations byStandard Occupational Classification (SOC) code. Posted medians: PivotHop corpus, 77,443 mapped postings, July 2026, cells floored at 100 postings. The salary-transfer correlation is Pearson’s r over the 123 occupations clearing that floor. Observed destination flows: see the source box in “Where people actually go.”