Why People Stay in Bad Relationships: The Psychology of Sunk Cost Thinking

From the outside, it seems so clear. The relationship isn't working. The patterns are repeating. Anyone watching can see that leaving would be the obvious move. And yet the person inside it stays — often for months or years after the problems became undeniable — doing the painful calculus of whether to go. It's tempting to attribute this to low self-worth, or fear, or simply not seeing what's happening. Those factors play a role. But there's a cognitive dimension that gets less attention: sunk cost thinking, and the way it quietly shapes decisions we believe we're making freely.

What Sunk Cost Thinking Actually Is

The sunk cost fallacy describes the tendency to continue investing in something because of what you've already put in, even when the current and future outlook doesn't justify it. Economically, this is irrational — past investment is gone regardless of what you do next. The question should be: what happens from here? But human decision-making isn't purely economic. We feel the weight of what we've spent. Leaving something you've invested years in isn't just making a practical calculation. It's confronting the meaning of all that time. In relationships, this plays out constantly. "We've been together seven years." "I gave up so much for this." "I can't let all of that be for nothing." These aren't just rationalizations. They're the actual experience of someone whose identity and history are now tangled up in a relationship that's stopped serving them.

The Investment Trap

The longer a relationship continues, the more it accumulates. Shared friends, shared history, interwoven finances, the habits and rituals that become the texture of daily life. Leaving doesn't just mean losing the relationship — it means dismantling a life that took years to build. Researchers at Northwestern University studying relationship persistence found that people were more likely to stay in unsatisfying relationships when the practical costs of leaving were high, even when they rated their relationship quality as poor. The investment itself — financial entanglement, shared housing, social network overlap — predicted staying more strongly than relationship satisfaction did. This is uncomfortable data, because it suggests that many people are not staying for love. They're staying because the cost of leaving feels too high to absorb.

Hope as a Mechanism

Alongside sunk cost thinking, there's the forward-looking counterpart: the belief that things will change. This isn't irrational either. Relationships do change. People do grow. Hard periods do end. The trap is when hope gets decoupled from evidence — when the belief that things will get better persists through repeated cycles of disappointment without requiring any update. This hope-without-evidence pattern is particularly powerful in intermittent reinforcement dynamics, where good periods and bad periods alternate unpredictably. The variable reward schedule this creates is neurologically similar to what makes gambling addictive. The good moments feel disproportionately good partly because they interrupt the bad.

The Tangent Worth Taking: Identity and the Relationship Self

People don't just invest time and resources in relationships — they invest identity. The person you were in a relationship, the shared future you imagined, the self-concept organized around "us" — these aren't minor attachments. When the relationship ends, those versions of yourself end too. That's a form of grief that's real and valid, and that often gets confused with grief for the actual person. "I miss who I was with them" is sometimes a more accurate statement than "I miss them." Separating those threads matters for making clear-eyed decisions.

What Helps

Research from the University of Toronto suggests that people facing difficult stay-or-leave decisions benefited significantly from what's called "self-distancing" — imagining the situation from the outside, or considering what advice they'd give to a close friend in the same circumstances. This works because it sidesteps the sunk cost trap somewhat. When advising someone else, you aren't carrying the weight of their history. You can assess the current reality more clearly. You can ask: given what's actually happening, what makes sense from here? The question is never "was this a good relationship to have been in?" It was. It can be fully that and still be time to leave.

The Permission Nobody Gives

Sometimes people stay because nobody has given them explicit permission to go — to say that the years they spent were real even if the relationship ends, that leaving doesn't retroactively invalidate the love, that the sunk cost is allowed to stay sunk. The past can be worth something without obligating the future. That's a distinction that takes time to feel, but it changes everything once it does.