"I know roughly where I am"

Most people who don't look would say they have a fair idea of their position. They know what they earn. They know the big outgoings. They haven't bought anything mad. There's probably a few hundred in the account. That's enough to go on.

The rough idea has two weaknesses, and they both push it in the same direction.

The first is that it's out of date. Your idea of what things cost was formed the last time you looked properly, which may be a year or two ago. Since then the insurance has gone up by sixty-four pounds and the broadband deal has ended and reverted to the standard rate. The streaming service put its price up twice. The gym you stopped going to is still taking its money. Each change arrived in a letter or an email, and each went in the drawer or the unread folder. None of them is large. Together they can easily amount to fifty or a hundred pounds a month that your rough idea doesn't include. Those figures are an illustration and I'm not offering them as a statistic, but you can check your own.

The second weakness is that the rough idea is kind to you. Its job is to let you feel all right, which is the whole reason you prefer it to the number. So where it's uncertain, it rounds in your favour. You remember the month you spent little and forget the weekend away. You think of your pay as it is before the deductions you don't think about.

Put together, a flattering estimate that's a year out of date will reliably tell you that you have more than you have. That's how a person with a perfectly adequate income ends up with a declined card in the third week of the month. They weren't reckless. They were steering by a map that no longer matched the road.

There's strong evidence that looking, by itself, helps. Benjamin Harkin and colleagues pooled 138 experiments, with nearly twenty thousand participants, in which some people were prompted to monitor their progress towards a goal and others weren't. The goals varied from study to study. Prompting people to monitor made them monitor much more, as you'd expect. It also made them more likely to reach the goal, with a moderate effect across the studies. The effect was larger when the information was physically recorded, meaning written down or entered somewhere. It was larger again when the results were reported to someone else or made public.

The reason isn't mysterious. You can't correct what you haven't noticed. If you see on the 9th that the month is running hot, you have three weeks to ease off, and easing off a little for three weeks is painless. If you find out on the 23rd at a checkout, there's nothing left to do but reach for the credit card. It's the same overspend, but in one case you can correct it and in the other it's become a small crisis. All that differs is when you found out.

Looking also turns up things in your favour. If you never read your statements, you never see the subscription you forgot to cancel, the direct debit for a service you no longer use, or the payment taken twice. You don't see the refund that was promised and never arrived, or the transaction you don't recognise at all. Banks ask customers to report anything unrecognised promptly, and the sooner it's raised, the simpler it generally is to sort out. A person who looks weekly catches these within days. A person who looks yearly may never catch them.

So the rough idea does harm. It's likely to be wrong in your disfavour by an amount that matters. It also switches off the one mechanism, which is noticing, that would put it right.