Pay Yourself First

Saving whatever's left at the end of the month is a plan that fails almost every month — not because you're bad with money, but because there is reliably nothing left.

Last verified 2026-08-02

Reverse the order and the problem mostly disappears

Most people run the month in this order: income arrives, spending happens, and savings gets whatever survives. The trouble is that spending is elastic. It quietly expands to fill whatever is sitting in the account, so 'whatever survives' converges on nothing regardless of how much came in.

Paying yourself first inverts it. The transfer happens on payday, before any spending decision gets made, and you live on the remainder. Nothing about the arithmetic changed — the same dollars go to the same places — but you've moved the decision to the one moment when the money is definitely there.

This is why raises so often fail to turn into savings. If the order stays the same, a bigger number at the top just gives spending more room to expand into.

What 'automatically' actually means in practice

Two ways to set it up, and either works. Split your direct deposit at the payroll end, so a fixed amount lands in savings and never touches checking. Or set a standing transfer at your bank, dated the day after payday rather than the same day, so it can't bounce if the deposit posts late.

Send it somewhere slightly inconvenient. A separate bank means the money takes a day to come back, and that day is usually enough for a want to stop feeling like a need. Same account, different bank — that one design choice does most of the work.

Then size it honestly. A transfer you cancel in month three taught you nothing except that you can cancel it. Start at a number you're confident about, even a small one, and let the next raise carry the increase.

Tying increases to raises, so nothing ever feels lost

Thaler and Benartzi's 'Save More Tomorrow' programme pushed the idea one step further: participants pre-committed to raising their savings rate at each future pay rise. Reported savings rates went from around 3.5% to about 13.6% over roughly 40 months, and most people stayed enrolled through four raises.

The design is the clever part. Because the increase comes out of money you never had in hand, take-home pay never drops. Nothing registers as a loss, and loss is the thing we resist hardest — a savings increase that reduces your current paycheck is the one people cancel.

The idea is now largely law: new 401(k) and 403(b) plans generally have to enrol employees automatically and step the rate up about one point a year. If your workplace plan does this, the mechanism in this lesson is already running on your behalf; your job is to not opt out of it.