The pricing problem is where most of the damage happens, and it runs in a predictable direction. Someone weighing an offer typically compares the total unvested balance on their equity statement against the raise on the table. Both numbers are wrong for the decision, and they are wrong in ways that push toward staying.
The unvested balance is a gross, uncertain, future figure. It is stated before tax, and equity delivered as compensation is ordinary income when it arrives, so the after-tax amount is materially smaller than the number on the screen. It is also contingent twice over: on the employee still being employed when each tranche vests, which is not entirely within their control, and on the share price at that future date, which is not within anyone's. A number that requires two future conditions is not comparable to a salary increase that starts next month.
The decision is rarely "all of it or none of it". Vesting arrives in pieces. The real cost of leaving today rather than after the next tranche is that tranche, not the entire unvested balance, and the cost of waiting for it is the delay to everything the new role pays. Framed that way, the answer is usually obvious in one direction or the other, and it changes every few months as the schedule advances.
A competing offer that carries its own grant replaces most of what is forfeited. An employer making an offer knows what it is asking a candidate to walk away from, which is why a joining grant is often part of the package. Where there is one, the honest comparison is not the forfeited grant against zero. It is the forfeited grant against the new grant plus the pay difference, over the same number of years. Doing that arithmetic is what turns a feeling into a number.
The behavioral half is loss aversion working on something that was never owned. Unvested equity shows up on a statement with a dollar value beside it, which is enough for most people to file it mentally as theirs. Leaving then registers as losing money rather than as declining a future gain, and loss aversion, the tendency to weigh a loss more heavily than an equivalent gain, does the rest. The schedule design compounds it: on a rolling grant there is always another cliff a few months out, so "just until the next vest" is a reason that never runs out. Anyone who has told themselves that twice about the same job is looking at the mechanism working exactly as intended.
Staying has a cost that does not appear on either side of the comparison. A household whose salary, unvested grants, already-vested shares and sometimes retirement plan balance all depend on one company is exposed to that company in four ways at once. A bad year there can remove the job and the value of the equity in the same quarter. Every additional year spent waiting for a vest adds to that concentration rather than reducing it, which is the part of the trade a purely financial comparison of two offers leaves out.
The arrangement is not sinister, and reading it that way leads to bad decisions. Deferred, forfeitable pay is how employers buy continuity, and it is also how an employee is paid more than a salary alone would deliver. The problem is not that the handcuffs exist. It is that they are usually valued by feel rather than by arithmetic, and feel is systematically biased toward staying.
