Three to six months of expenses. It is the most repeated number in personal finance, it appears in almost every guide, and nobody can tell you where it came from. It is not the output of a study. It is a convention that hardened.
Which would be harmless if it were only imprecise. The problem is who it is addressed to.
The gap it is spoken into
Every year the Federal Reserve asks a nationally representative sample of American adults whether they could cover a $400 emergency expense using cash or its equivalent. In the 2024 survey, published in May 2025, 63% said they could.
37%
of US adults could not cover a $400 emergency from cash — they would borrow, sell something, or not pay it
Hold those two numbers next to each other. Nearly four in ten adults cannot produce four hundred dollars, and the standard advice tells them to accumulate several months of living costs. For a household spending $3,000 a month, three to six months is nine to eighteen thousand.
That is not a stretch target. It is between twenty and forty-five times the amount they currently cannot find, and a goal at that distance does not motivate. It gives permission not to begin, because beginning would be pointless.
What the volatility research actually measures
There is a body of work that asks the empirical version of the question — not “how much would feel safe” but “how much would have been enough”, measured against what actually happens to household finances.
The JPMorgan Chase Institute has run it across millions of accounts. Family income and spending each fluctuate by roughly 30% month to month, and for middle-income households the resulting unplanned expense comes to about $1,300. Their estimate of the buffer families need to absorb that volatility is roughly six weeks of take-home pay — peaking at about $2,400 for middle-income families in their late thirties and early forties.
6 weeks of pay
the buffer measured against real household volatility — against a convention of three to six months of expenses
Six weeks of pay is not the same quantity as three to six months of expenses, and it is considerably smaller. The convention is somewhere between two and four times the measured requirement.
What these numbers are
The Fed figure is a survey response about a hypothetical $400 expense — a deliberately specific probe, not a measure of total savings. The volatility estimates come from observed account data, which is stronger evidence about what happens and says nothing about what any individual household should hold.
Neither is a rule. What they establish jointly is narrower and more useful: the shocks households actually absorb are in the hundreds to low thousands, and the conventional target is calibrated to something else.
Why the larger number is not the safer one
It seems intuitively cautious to name a big target. If someone saves less than you told them to, they still end up with something.
That is not how goals of this shape behave. A target far enough away to feel unreachable does not produce partial effort; it produces abandonment, and it competes for attention with things that feel achievable. The household that cannot find $400 and is told to find $12,000 does not save $3,000. It concludes that this is not for them.
Meanwhile the first few hundred dollars is where most of the protection lives, because that is the size of the events that do the damage. The car repair. The vet. The excess. The week of lost shifts. Almost nothing in ordinary financial life is destroyed by a shortfall of nine thousand dollars; a great deal is destroyed by a shortfall of four hundred.
- Make the first target $500, and treat reaching it as the whole goal. Not a milestone on the way to a bigger number — a finished thing. It covers the modal emergency and it is reachable, which the other target is not.
- Then aim at your own worst realistic month, not at a multiple of your rent. What has actually gone wrong in the last three years, and what did it cost? That number is yours and it is usually smaller than the convention.
- Count in weeks of pay, not months of expenses. The measured requirement is expressed that way for a reason: pay is what refills the buffer, and expenses are what drains it.
- Keep it somewhere boring and reachable. A buffer in an account you cannot access within a day is not a buffer; it is an investment with a story attached.
- Stop at the number and go and do something else. Money past the buffer has better jobs — debt, pension, the thing you actually want. Over-saving into cash is a real cost, quietly paid.
The first of those is the only one that needs doing this month, and it is deliberately small.
Try this
Write down every unexpected expense that has actually hit you in the past three years, with what each one cost. Not what you fear. What happened.
Most people produce a list of items between one hundred and two thousand dollars, and are surprised by how few there are. That list is a far better specification for a buffer than any multiple of monthly outgoings — and for many people it argues for a smaller target than the one they have been failing to reach.
The parts of this that are genuinely hard
For a substantial share of the households in that 37%, the shortfall is not a savings-behaviour problem. It is an income problem, and no target-setting advice touches it. Telling someone whose outgoings exceed their income to aim at $500 rather than $12,000 has made the advice honest, not achievable, and the honest thing is to say so rather than to imply the gap is one of discipline.
The convention is also not wrong for everybody. If your income is irregular — self-employment, commission, seasonal work — or if a household runs on one salary, the risk is not a $400 shock but a three-month gap, and months of expenses is exactly the right unit. The error is not the number. It is applying one number to every shape of risk.
And a buffer does one job. It does not fix insurance, a pension, or a debt at 24%. It buys the ability to meet a small disaster without borrowing, which is the specific thing four in ten adults currently cannot do.
The bottom line
The measured requirement is around six weeks of pay, the shocks that actually arrive cost hundreds rather than thousands, and the conventional target is several times larger than either.
So make the first goal $500 and let it be finished. It is not a lesser version of the real advice. On the evidence it is closer to the real advice than the number you have been given.
Sources (3)
Every factual claim in this article comes from one of these sources.
Keep exploring
Continue reading
Deeper into what you just read.
Explore more
Adjacent ground, if this was useful.
One thing worth knowing
A surprising finding. The evidence behind it. And what it means for your life. BetterYou+ sends one useful idea at a time—nothing to keep up with.