Why Do I Keep Paying for Subscriptions I Never Use? Loss Aversion, Explained

Loss aversion is the reason: cancelling feels like giving something up, and giving something up registers more strongly than the equivalent gain of keeping the money. So the monthly charge quietly loses to the discomfort of the cancel button, every month, for years.

That is the short answer. The longer one is more interesting, because the famous version of loss aversion — "losses hurt about twice as much as gains feel good" — turns out to be shakier than almost everything written about it admits. A 2024 meta-analysis of the actual experiments puts the effect at roughly 1.3 times, not 2.25, and in most of the datasets it examined the result was statistically compatible with no loss aversion at all.

Both things can be true. The pattern is real in your subscriptions. The number people quote at you is probably wrong.

What's happening psychologically

Three things are stacked on top of each other, and they are worth separating because only one of them is loss aversion proper.

1. The cancellation is framed as a loss. You are not choosing between £9 and a service. You are choosing between keeping something you already have and not having it. Prospect theory's core claim is that we evaluate outcomes as changes from where we are now, not as absolute states — and a change framed as a removal carries more weight than the same change framed as an addition. The £9 you would keep is abstract. The thing you would no longer have is specific.

2. You already own it, which makes it feel worth more. Ownership itself inflates value. The service you have had for fourteen months is not being compared to a service you might buy today; it is being compared to the version of yourself who has it.

3. The month you have already paid for pulls on you too. "I've had it since January" is a sunk cost, and sunk costs should be irrelevant to a forward-looking decision. They are not irrelevant to how the decision feels. This is a different bias from loss aversion, frequently mistaken for it, and it is the one most likely to be doing the work when the subscription is annual.

Add a fourth, practical layer: the cancel flow is usually designed to be slightly worse than the sign-up flow. That is not psychology, it is architecture — but it borrows its power from the three biases above, because a small friction placed in front of a decision that already feels like a loss is enough to stop it entirely.

What research actually says

Here is where this article is going to be less satisfying than most, and more accurate.

The classic figure comes from Tversky and Kahneman's 1992 work, which estimated a loss-aversion coefficient — usually written λ — of 2.25. That is the source of "losses hurt twice as much", and it has been repeated for thirty years.

In 2024, Walasek, Mullett and Stewart published a meta-analysis in the Journal of Economic Psychology that pooled the available risky-choice datasets: 19 datasets from 17 published articles, 1,072 participants in total, with individual sample sizes from 16 to 235. Their result:

> A pooled λ of 1.31, with a 95% confidence interval of [1.10, 1.53].

And two details that matter more than the headline:

- The confidence intervals included loss neutrality — no loss aversion at all — in 12 of the 19 datasets.
- The confidence intervals included the classic value of 2.25 in only 6 of the 19.

The authors also report "a considerable amount of both within- and between-study heterogeneity", and say they could not run moderator analyses because the designs varied too much.

So the fair summary is this: averaged across these experiments, losses appear to weigh somewhat more than gains, by maybe 30% rather than 125%, with a lot of variation between people and studies, and with a real possibility that for many individuals the asymmetry is small or absent.

An everyday example

A streaming service, £8.99 a month, signed up to in February for one series.

Each month the decision presents itself for about four seconds, usually when the payment notification arrives. In those four seconds the account holder is not comparing £8.99 against zero. They are comparing two futures: one where the library is still there and one where it is not. The library is vivid, the £8.99 is a rounding error, and there is a half-formed thought that they will "definitely watch something at the weekend".

Eighteen months later that is £161.82, and the series finished in March.

What makes this example useful is what it is not. It is not a failure of arithmetic — anyone can multiply 8.99 by 18. It is that the arithmetic never gets run, because the decision is settled by feel before the number is reached. The reason the annual audit works is not that it teaches you maths. It is that it changes the framing from "cancel this" to "choose this again".

What this does not necessarily mean

This is the section I would most like you to read.

- It does not mean you are bad with money. The same pattern shows up in people who are careful with money everywhere else, because it is a framing effect, not a character trait.
- It does not mean loss aversion is a fixed quantity inside you. On the best current pooled estimate the average effect is modest and varies enormously between people and situations. If someone tells you your losses hurt exactly 2.25 times as much as your gains feel good, they are quoting a thirty-year-old parameter that a 2024 meta-analysis did not reproduce.
- It does not mean loss aversion is a myth either. The pooled estimate is above 1, and the lower bound of the interval, 1.10, is still above neutrality. "Smaller than advertised" is not the same as "not there".
- It is not a diagnosis of anything. Keeping unused subscriptions is an ordinary pattern in ordinary brains. It is not a symptom, and nothing here is a basis for concluding anything clinical about you or anyone else.
- It does not make the companies villains or you a dupe. Friction in a cancel flow is a design decision; your response to it is a normal human response. Both can be true without anyone needing to be stupid or evil.

Practical things to try

The useful moves all work by changing the framing rather than by trying harder.

1. Make it an opt-in, not an opt-out. Once a year, list every recurring charge and ask of each one: would I sign up for this today, at this price, knowing what I know? A "no" is now a decision not to buy, which is psychologically much cheaper than a decision to cancel.
2. Read the bank statement, not your memory. Export three months of transactions and sort by merchant. The subscriptions you have forgotten are, by definition, the ones you will not remember to audit.
3. Price it in years. £8.99 is noise. £107.88 is a decision. The multiplication is the intervention.
4. Put the friction on the other side. Set a calendar reminder two days before each annual renewal rather than deciding at the moment of the charge. A decision made in advance is made by a different, less loss-averse version of you.
5. Cancel in pairs. Doing two at once converts "I am losing something" into "I am doing an admin task", which is a frame with no loss in it.
6. Use the pause, where it exists. Several services offer a pause rather than a cancel. If the barrier is the feeling of giving something up, a pause removes the feeling without the charge.
7. Separate the sunk cost explicitly, out loud. "I have paid for this for fourteen months" is a fact about the past. Say it, then say "and that is not a reason", then decide. Naming a bias is weak medicine, but it is reliably better than nothing.

What none of these do is make the discomfort disappear. They make the decision happen anyway.

When professional help may be appropriate

For the ordinary version of this — a few forgotten subscriptions — no help is needed beyond an afternoon with a bank statement.

It is worth talking to someone if money decisions are causing persistent distress, if avoidance has spread beyond subscriptions to opening post or checking balances, or if spending feels out of your control rather than merely untidy. Free debt advice services exist in most countries and are the right first call for the financial side; a GP is the right first call if the distress is the bigger part. Our Mind Decoded writes about patterns in everyday behaviour, and that is a different thing from advice about your situation — this article is not financial advice and not a clinical assessment.

Frequently asked questions

Q: What is loss aversion, in plain English?
A: It is the tendency to weigh a loss more heavily than a gain of the same size. Losing £10 registers as a bigger event than finding £10, even though the amounts match. The important nuance, which most explanations skip, is the size of the gap: the classic figure of 2.25 times comes from 1992, while a 2024 meta-analysis of 19 datasets and 1,072 participants puts the pooled estimate at 1.31, with a confidence interval of 1.10 to 1.53. Real, but considerably smaller than its reputation.

Q: Is loss aversion the same as the sunk cost fallacy?
A: No, though they often appear together and get confused. Loss aversion is about how you weigh a future change — giving something up feels worse than the matching gain feels good. The sunk cost fallacy is about the past — continuing with something because of what you have already spent on it, which logically should not affect the decision at all. With a monthly subscription, loss aversion is usually doing most of the work. With an annual one, or a course you paid for upfront, sunk cost tends to take over.

Q: Why is cancelling a subscription so much harder than signing up?
A: Partly psychology and partly design. The psychological part is that signing up is framed as a gain and cancelling as a loss, and those are not processed symmetrically. The design part is that many sign-up flows are one click while cancel flows require logging in, finding a settings page and confirming twice. Neither alone would stop you. Together, a small friction placed in front of a decision that already feels like a loss is often enough to postpone it for another month, and then another.

Q: Does loss aversion theory apply to everyone equally?
A: On current evidence, clearly not. The 2024 meta-analysis found "a considerable amount of both within- and between-study heterogeneity", and its confidence intervals were compatible with loss neutrality — no asymmetry at all — in 12 of the 19 datasets examined. The honest position is that loss aversion is an average tendency across groups of people in particular kinds of decisions, not a constant that can be read off for an individual. Anyone selling you a personal loss-aversion score is overreaching.

Q: How do I stop loss aversion affecting my money decisions?
A: You mostly cannot stop the feeling, so change the question instead. The single most effective move is to convert every renewal into a fresh purchase decision: not "should I cancel this?" but "would I buy this today at this price?" That reframing removes the loss from the equation, because declining to buy something does not feel like losing it. Pricing the subscription in years rather than months, and deciding two days before the charge rather than on the day, work the same way.

Q: Why does the "losses hurt twice as much as gains feel good" figure keep getting repeated?
A: Because it is memorable, it comes from genuinely important work, and it was the best available estimate for a long time. Tversky and Kahneman's 1992 λ of 2.25 entered popular writing and has been copied forward ever since, largely without being rechecked. The 2024 meta-analysis is one of the first attempts to pool the underlying datasets, and it did not reproduce 2.25 — its intervals included that value in only 6 of 19 datasets. Our Mind Decoded's view is that a pattern being real and a number being right are two separate claims, and the second one is usually the weaker.

Q: Is keeping unused subscriptions a sign of a bigger problem?
A: Usually it is a sign of nothing at all beyond a busy life and a well-designed cancel flow. It is one of the most common money behaviours there is, and it shows up in people who are otherwise meticulous. It would be worth more attention if it sat alongside avoiding your bank balance, leaving post unopened, or spending that feels out of your control rather than simply unexamined — and in that case the useful first step is free debt advice or a GP, not a psychology article.


Take the quiz: which bias is keeping your subscriptions alive?

Six questions, two minutes. It separates loss aversion from sunk cost from plain inertia, and tells you which of the seven moves above is the one that will actually work for you.

Subscribe to our YouTube channel: (71) Mind decoded - YouTube

Connect

Questions or thoughts? Reach out anytime.

Email

subscribe to our newsletter

hello@our-minddecoded.com

© 2026. All rights reserved.

Terms of Service Privacy Policy Disclaimer