Key Takeaways
- The years you’ve already invested are a sunk cost. Research shows people reliably continue a course of action because of what they’ve already spent, which is not a reason.
- People systematically overestimate how much a change will affect them, in both directions, so your forecast about leaving is probably wrong in a predictable way.
- Actions produce more regret in the short term, but inactions produce more regret in the long run. The regret you’re bracing for isn’t the one that lasts.
- Job satisfaction reliably spikes after a voluntary move and then declines, which is worth knowing before you read the spike as proof.
- A better question than “should I quit” is whether the specific thing that’s wrong can change where you are.
It’s about eleven at night and you have a document open with two columns in it. You’ve made this document before, possibly several times, and you may still have the earlier versions. The columns fill up. Neither side wins. At some point you close the laptop having decided, once again, to think about it properly at the weekend, and you carry the whole thing around for another fortnight until something at work makes it urgent again for an evening.
The reason the list isn’t working is that this was never an information problem. You already know what the job is like. What you’re doing is trying to predict how a future version of you will feel about a decision this version has to make with incomplete information, under pressure, while tired. Psychology has surprisingly specific things to say about why that goes wrong, and about what to ask instead.
The list isn’t failing because you haven’t thought hard enough
Pros and cons assume the difficulty is weighing known quantities. It isn’t. Three things are quietly distorting the exercise, and none of them get better with more thinking.
The first is that you’re counting things that shouldn’t count. The second is that your forecast about life after the decision is unreliable in a known direction. The third is that you’re afraid of the wrong regret.
Take them in turn, because each one changes what belongs in the columns.
Distortion one: the years you’ve already spent
Hal Arkes and Catherine Blumer demonstrated what they called the sunk cost effect: a greater tendency to continue an endeavour once an investment of money, effort or time has been made, even when continuing is no longer the better option (Arkes & Blumer, 1985).
It shows up in this decision constantly, usually dressed as prudence. Six years in this industry. The training that took two years to complete. The reputation you built. The fact that leaving would “waste” all of it.
Here’s the thing about those years: they’re gone regardless. Staying doesn’t recover them. The only question a decision can answer is what happens from here, and the six years are identically spent in both futures. They feel like they belong in the ledger because they were expensive, and expense is exactly what makes sunk costs so persuasive.
The test is simple. If you’d arrived at this exact job yesterday, with no history, knowing what you know now, would you take it? If the answer is no, the years aren’t an argument for staying. They’re an explanation for why leaving feels harder than it is.
Distortion two: your forecast is wrong in a predictable direction
You’re not just deciding. You’re predicting: how you’ll feel in the new place, how you’ll feel if you stay, how much either will matter in a year.
Timothy Wilson and Daniel Gilbert’s work on affective forecasting found people are consistently poor at this in a specific way. We overestimate the intensity and duration of our emotional reactions to future events, a pattern they called the impact bias, largely because we focus narrowly on the change itself and underestimate how quickly we adapt and how much else will be going on (Wilson & Gilbert, 2005).
Applied here, that cuts both ways, which is the useful part. The relief you’re imagining after handing in your notice is probably smaller and shorter than you picture. So is the catastrophe you imagine if it goes badly. Both forecasts are inflated by the same mechanism.
That’s not an argument for staying or going. It’s an argument for weighting your predictions less heavily than they feel like they deserve, and for weighting the concrete and verifiable more.
Distortion three: you’re bracing for the wrong regret
Most people making this decision are primarily managing fear of one regret: leaving and wishing they hadn’t.
Thomas Gilovich and Victoria Medvec examined the structure of regret and found a clear temporal pattern. Actions, errors of commission, generate more regret in the short term. Inactions, errors of omission, produce more regret in the long run (Gilovich & Medvec, 1995).
Sit with the asymmetry, because it matters for a decision you’ll be living with for years. The regret you’re most afraid of, the one attached to leaving, is the kind that tends to fade. The regret attached to not leaving is the kind that tends to grow.
That isn’t a licence to quit, and this finding doesn’t say action is always right. It says the fear you’re weighting most heavily is attached to the shorter-lived of the two, which is worth knowing when the fear is doing most of the deciding.
What tends to happen after people leave
Now the part that gets left out of most writing on this, because it complicates the happy ending.
Wendy Boswell, John Boudreau and Jonathan Tichy tracked managers’ job satisfaction around voluntary job changes. The pattern was consistent enough to name: low satisfaction precedes a voluntary move, satisfaction rises sharply immediately after it, and then it declines. They called it the honeymoon-hangover effect (Boswell, Boudreau, & Tichy, 2005).
Three things follow.
The dip before you leave is normal and it’s part of why you’re reading this. The lift afterwards is real, and it’s also partly the move itself rather than the destination. And the decline that follows doesn’t mean the new job was a mistake, any more than the honeymoon meant it was perfect.
The practical implication is about timing your judgment. Don’t evaluate the new role at week three, when the honeymoon is doing the talking, and don’t panic at month nine when it isn’t. And if you’ve changed jobs several times and each one has followed the same arc, that pattern is information about something other than the individual jobs.
A better question than “should I quit”
Here’s the reframe that makes this tractable.
“Should I quit” is a question about a job. It’s too big to answer, which is why the list never resolves. The more answerable version is: what specifically is wrong, and can that specific thing change where I am?
That splits one impossible question into two smaller ones with real answers. Workload can sometimes be renegotiated. A manager can sometimes be changed, or outlasted. A values mismatch between you and what the organisation rewards usually cannot, because it isn’t a policy, it’s the organisation.
Naming the specific thing also tells you what a good outcome looks like, which the two-column list never does. “Leave” isn’t a goal. “Stop being responsible for something I have no authority over” is one, and it might be reachable without leaving. Our guide to recovering from burnout without quitting walks through where those levers sit, and it’s worth reading before you conclude there aren’t any.
The third option people skip
The decision usually gets framed as binary: stay as you are, or leave. There’s a middle that most people never seriously price.
Roles are more negotiable than they look from inside them, particularly for someone who is already competent and whom the organisation would rather not lose. Changing team, changing manager, dropping a responsibility, moving sideways, going to four days, taking a defined period of leave. Each of these is smaller than resigning and each produces real information about whether the place can accommodate you.
People skip this step for a reason worth naming. Asking is exposing. It reveals that you’re unhappy, which can feel like handing someone leverage, and there’s a real fear that raising it marks you as a flight risk. Sometimes that fear is accurate. Often it’s a prediction being treated as a fact, and it’s the same overconfident forecasting the impact bias describes.
The cost asymmetry is worth doing out loud. If you ask and the answer is no, you’ve lost some comfort and gained a genuine piece of evidence: this place will not change for you. That’s exactly what you needed to know, and you now have it without having resigned to get it. If you don’t ask, you’ll be deciding on a guess about what they would have said.
Our piece on the psychology of career transitions covers the two directions adjustment can run, changing yourself to fit a role or changing the role to fit you, and which one your particular job actually allows.
Four questions worth answering before you decide
1. If I arrived here yesterday, would I take this job? This strips out sunk cost. It won’t give you the answer, but it tells you how much of your reluctance is about the past rather than the future.
2. What exactly is wrong, in one sentence, without the word “everything”? If you can’t name it, that’s information too. Sometimes the problem isn’t the job. Sometimes it’s that this period of your life is hard and work is the most visible part of it.
3. Has anything I’ve tried actually been tried? Not thought about. Tried. Asking for the change, having the conversation, applying for the internal move. People often decide a place can’t change without testing it, because testing it feels riskier than concluding it.
4. What’s the smallest version of this decision? Quitting is the largest available move, and it isn’t always the first one. A conversation with your manager, a role change, a period of leave, a reduced week: each is reversible in ways resignation isn’t, and each generates real information rather than predicted information.
When it isn’t a preference, it’s health
Everything above assumes a decision between tolerable options. Sometimes that’s not the situation, and the framework changes.
If the work is affecting your health, if you’re experiencing sustained anxiety or low mood that tracks clearly with the job, if there’s bullying, harassment or discrimination, or if the environment is unsafe, then this stops being an optimisation problem. Enduring it is not the neutral, prudent option it can feel like from inside.
Burnout is worth naming precisely here. The World Health Organization classifies it as an occupational phenomenon resulting from chronic workplace stress that hasn’t been successfully managed, and explicitly not as a medical condition. That framing matters for this decision, because it locates the problem in the relationship between you and the conditions rather than in you.
If you’re not sure whether what you’re experiencing is dissatisfaction or something heavier, our guide to work anxiety covers the difference and where the line sits.
When to talk to a professional
Worth speaking to someone if you’ve been circling this decision for more than a few months without moving, if the job is affecting your sleep, your health or your relationships, if you notice you’re staying mainly because leaving feels frightening rather than because staying feels right, or if the decision has become the thing you think about instead of the thing you act on.
A therapist isn’t there to tell you whether to quit. What they can do is help you see which parts of the paralysis belong to the job and which belong to older patterns about risk, security and what you think you deserve. That distinction is genuinely hard to make alone, and it’s usually the thing keeping the document open.
The takeaway
You probably aren’t stuck because you lack information. You’re stuck because the decision is being made by a system that overweights what you’ve already spent, mispredicts how you’ll feel, and fears the regret that fades over the regret that lasts.
None of that tells you what to do. What it does is clean up the inputs. Take the sunk years out of the ledger. Hold your forecasts loosely in both directions. Notice that leaving and staying both carry regret, and that they don’t age the same way. Then swap the unanswerable question for the answerable one: what specifically is wrong, and can it change here?
Whatever you land on, it isn’t permanent. People change careers more than once now, and the version of you making this call has considerably more room to revise it than the eleven-o’clock version believes. Our piece on why career transitions feel so destabilising covers what the other side of the decision actually feels like, which is worth knowing before you get there.
If the decision was taken out of your hands, the psychology looks different. Our piece on the mental health impact of losing a job covers that situation directly.
If the job you’re thinking of leaving is one you’ve done for most of your working life, our guide to a midlife career change looks at the extra weight that decision tends to carry.
If you’ve been circling this for months
Being stuck on a decision this size is exhausting in its own right, and it often has less to do with the job than with what leaving would mean. Therapy is useful for exactly this: separating the practical question from the older beliefs about safety and worth that are making it unanswerable. Our guide to finding the right therapist covers what to look for and how to begin.
Frequently asked questions
How do I know when it’s time to quit my job?
There’s no research-validated threshold, and anyone offering one is guessing. What the evidence supports is a better question: name the specific thing that’s wrong, then ask whether it can change where you are. Workload and management sometimes can. A fundamental values mismatch usually can’t.
Should I stay because of how much time I’ve already invested?
No. That’s the sunk cost effect, a documented tendency to continue something because of prior investment of money, time or effort. Those years are spent whether you stay or go, so they can’t distinguish between the two futures. Ask instead whether you’d take this job today.
Will I regret leaving my job?
Possibly, in the short term. Research on the structure of regret found that actions generate more regret soon afterwards, while inactions generate more regret over the long run. So the regret attached to leaving tends to fade, and the regret attached to staying tends to accumulate.
Why did I feel great in my new job and then worse?
That’s the honeymoon-hangover effect. Studies tracking managers through voluntary job changes found satisfaction is low before a move, spikes immediately after it, then declines. The dip doesn’t mean you chose wrong. It means the initial lift came partly from the change itself.
Is it burnout or do I just need a different job?
They overlap and they aren’t the same. The World Health Organization defines burnout as an occupational phenomenon arising from chronic workplace stress that hasn’t been successfully managed, which locates it in the fit between you and the conditions. That fit can sometimes be changed without changing employer.
Sources
- Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35, 124–140. Link
- Boswell, W. R., Boudreau, J. W., & Tichy, J. (2005). The relationship between employee job change and job satisfaction: The honeymoon-hangover effect. Journal of Applied Psychology, 90, 882–892. Link
- Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What, when, and why. Psychological Review, 102(2), 379–395. Link
- Wilson, T. D., & Gilbert, D. T. (2005). Affective forecasting: Knowing what to want. Current Directions in Psychological Science, 14(3), 131–134.
- World Health Organization. (2019). Burn-out an “occupational phenomenon”: International Classification of Diseases. Link
- National Institute of Mental Health. Anxiety Disorders. Link
This article is for educational and informational purposes only and is not a substitute for professional medical or mental health advice, diagnosis, or treatment. Always seek the guidance of a qualified mental health professional with any questions about your mental health or a medical condition.
