What the data actually shows
The most concrete reason is financial. A paycheck pays rent, and in the United States a job is also frequently the source of health insurance, which makes leaving riskier than the salary alone implies. 'Golden handcuffs' — high pay, vested benefits, accrued seniority, or equity that pays out over time — deliberately raise the cost of walking away, and they work as designed.
Several robust biases reinforce staying. Status-quo bias, documented by Samuelson and Zeckhauser, describes a general tendency to prefer leaving things as they are; the current arrangement is treated as a default, and any change is judged against it. Loss aversion makes the certain losses of leaving — income, security, familiar routines — loom larger than the uncertain gains of something new.
Sunk-cost thinking adds another layer. Years invested in a role, a company, or a profession feel like something that would be wasted by leaving, even though those years are already spent and cannot be recovered. Combined with a simple fear of the unknown, these biases mean that even people who report active dislike of their work often stay. Gallup's engagement surveys consistently find that only a minority of workers worldwide feel actively engaged, while a large share are disengaged yet remain in place.
Why this feels different from how it actually is
From the outside, staying in a hated job can look like a simple lack of courage, because the costs of the job are visible — the dread, the complaints — while the costs of leaving are not. The lost income, the gap in health coverage, the risk of a worse situation, the caregiving responsibilities that depend on a steady salary: these are real, but they are easy for observers to discount.
From the inside, the decision rarely feels free. It feels constrained by obligations to a family, a mortgage, or dependents, and by a job market that may not offer an obviously better alternative. The honest framing is that 'just leave' assumes options and a safety net that many people do not have.
There is also a strong parallel with relationships. The same combination of sunk cost ('I've put years into this'), loss aversion ('what if the next one is worse?'), and status-quo bias ('better the devil I know') that keeps people in relationships that are not working operates almost identically when the relationship is with an employer.
What the research says to do about it
Because the biases push in a predictable direction, the research-informed move is to make the comparison more honest rather than to simply 'be braver.' Sunk-cost reasoning is countered by asking what you would choose today if the past years were not a factor — they are already spent either way, so they should not weigh on the forward decision.
Loss aversion and status-quo bias are eased by making the alternative concrete instead of leaving it as a vague unknown. Researching actual options, building a financial buffer, and understanding the real continuity of benefits (such as how health coverage would carry over) shrinks the uncertainty that the brain is reacting to, which is often what was inflating the cost of leaving.
It also helps to separate the parts of the job you can change from the parts you cannot. Some dissatisfaction responds to negotiating responsibilities, hours, or team, which is lower-risk than a full exit; the rest may genuinely require leaving. The point is to base the decision on a clear-eyed reading of constraints, not on willpower alone.
What the research says does not help
Treating it as a willpower problem does not help and misreads the situation. Telling someone they are 'settling' ignores the financial obligations, benefits lock-in and limited options that often make staying the reasonable short-term choice. The barrier is usually structural and cognitive, not a deficit of grit.
Equally, dismissing the constraints as mere excuses does not help. The costs of leaving — lost income, a coverage gap, caregiving duties, a weak job market — are frequently real, and pretending they are not leads to advice that does not fit the person's actual life.
Impulsively quitting on a bad day, with no buffer or plan, tends to backfire, because the very loss aversion that kept someone stuck can flip into a rushed move that lands them somewhere worse. The biases are best handled by making the alternative concrete first, not by overriding them in a single frustrated decision.
Real numbers in context
Gallup's workplace research consistently finds that only a minority of workers worldwide feel actively engaged in their jobs, while a substantial share are not engaged or are actively disengaged. The precise percentages move year to year, so they are best read as a stable pattern: feeling disengaged at work is common, and so is staying anyway.
The strength of benefits lock-in is hard to put a single number on, but in the U.S. context the link between employment and health insurance is a well-documented reason people stay in jobs they would otherwise leave — sometimes described as 'job lock.' Combined with sunk-cost and status-quo effects from the behavioural-economics literature, the overall picture is that staying is over-determined: several independent forces all point the same way.