How Big Should an Emergency Fund Be? Worked With Your Own Numbers
Not three to six months of income. Three to six months of one stripped-down month, times a dial you score. The whole calculation, worked in public.
The honest answer is that an emergency fund is sized in two steps, and almost every version of the question skips the first one. Step one: find what one stripped-down month actually costs. Not your income, not your normal spending. Housing, utilities, food eaten at home, transport, insurance, minimum debt payments and the things children genuinely need. Step two: multiply that by a number between three and six that you score rather than feel. On the invented household below, the survival month is $4,000.00 and the scored dial is 4, so the full target is $16,000.00. The starter target is smaller and comes first.
Sizing off income is the common mistake, and it is why so many people quietly decide the whole idea is impossible.
Step one: the survival month is a subtraction
Seven lines, and nothing else. Everything optional pauses, and paused is not the same as cut.
| Line | Amount |
|---|---|
| Housing | $1,680.00 |
| Utilities and phones | $360.00 |
| Groceries, eating at home | $760.00 |
| Transport and fuel | $310.00 |
| Insurance | $410.00 |
| Debt minimums only | $290.00 |
| Children's essentials | $190.00 |
| Survival month | $4,000.00 |
Here is the part that surprises people. That $4,000.00 is 65.07% of this household's ordinary month of $6,146.84. The other $2,146.84 is the goal payment, the sinking funds, the fun band and the breathing room in every category. All of it pauses when trouble arrives. That gap is the entire reason emergencies are survivable at all, and it is measurable before anything goes wrong. If you have never built the underlying plan, a budget that survives a real month is where the $6,146.84 figure comes from.
Step two: score the dial instead of feeling it
Three to six is not a mood. Count the things that make a household steadier and the things that make it bumpier, then subtract.
Two earners in different industries. Predictable hours. A low insurance deductible. A newer vehicle. A landlord who handles repairs. Skills that are easy to re-employ. No dependants.
One earner, or two in the same industry. Hours that swing. A high deductible. An ageing house or car. Children. Health conditions in the household. Self-employment.
The practice household scores four bumpy against two steady, which is plus two, which sets the dial at 4 months. Your neighbor's three and your six can both be right, because you are not scoring the same household.
The worked example, end to end
| Step | Working | Result |
|---|---|---|
| 1. Price the likeliest shocks | five priced locally, tallest circled | $3,600.00 |
| 2. Strip the month | seven must-pay lines, minimums only | $4,000.00 |
| 3. Starter target | the larger of the two above | $4,000.00 |
| 4. Score the dial | four bumpy minus two steady | 4 months |
| 5. Full target | $4,000.00 times 4 | $16,000.00 |
Two targets, because they answer two different questions. The starter of $4,000.00 answers can this household survive one hit. The full floor of $16,000.00 answers can it survive a season with no income at all. Reaching the first one changes how the second one feels.
Why the dial matters more than it looks
Run the same survival month at the two ends of the range and the spread is not small. Three months is $12,000.00. Six months is $24,000.00. That is a difference of $12,000.00 on identical monthly costs, decided entirely by how bumpy the household is. This is why a single universal number was never going to work, and why anyone quoting one is describing a household that is not yours.
How long it takes, honestly
On the practice file the household has $1,022.84 a month available after the plan is set, so the full floor of $16,000.00 takes 16 months on autopilot. Sixteen months is a long time. It is also finite, which is a completely different feeling from open-ended, and it is why writing the number down matters more than the number itself.
Two things make it shorter. The first is that the starter target arrives long before the full one. The second is that a household clearing debt frees up payments as each balance dies, and that money has to go somewhere. Both payoff orders, run on the same balances, shows exactly how much comes free and when.
Where the money sits
The usual three-part test is: safe, earning something rather than nothing, and one to two days away by transfer. Fast enough for a real emergency, slow enough that an impulse loses its window. This page names no bank, no account and no product, and it will not, because the right container depends on facts about you that a page cannot see. What a page can do is give you the test to hold a container up against.
Pricing the shocks, which is the step people skip
Step one of the method above quietly contains a second job: pricing your five likeliest emergencies. Most households already know what their monsters are. The car. The furnace or the boiler. A tooth. The dog. A flight home. What almost nobody has done is put a local number beside each one, and ten minutes of real pricing turns five vague dreads into five numbers, which can be planned for.
On the practice file the tallest priced shock is a $3,600.00 transmission rebuild. That is the smallest emergency the fund must be able to eat in one bite, which is why it competes with the survival month for the starter target. If your tallest shock is bigger than one survival month, your starter target is the shock. If it is smaller, your starter target is the month.
Pricing is also where the exercise stops being frightening. A number you have looked up is a smaller thing than a number you have been avoiding, even when the number turns out to be larger than you guessed. And several of the five usually turn out to be insured, which is a different conversation with a different answer.
The trap of the average month
There is one more reason the survival month rather than a typical month is the right input. On the practice household, six months of real deposits average out to $6,350.01 while the lowest of the six is $6,146.84. That is a gap of $203.17, and four of the six months came in below the average. A fund sized off the average is a fund sized off a month that, in this household, mostly did not happen.
The same logic runs in the other direction on the spending side. Sizing an emergency fund off what a household normally spends bakes in everything that would stop the moment income stopped: the goal payment, the sinking funds, the fun. On the practice file that is $2,146.84 a month of things that pause. Including them would inflate the target by more than half, for no protective benefit at all.
If a month is unpredictable
Households paid hourly, seasonally or on commission have a harder version of this problem, because there is no single month to strip. The workable move is to build the survival month from the lines rather than from a past month, since rent and insurance do not care what the hours did. Pay that arrives every two weeks has its own arithmetic problem on top, and it is worth reading if your pay dates and your bill dates keep colliding.
One number, tonight, on paper. Almost everybody who does this finds it is smaller than the figure they had been quietly dreading, and a smaller number is a startable one.
Refilling, which is the half nobody plans
A fund that gets used is a fund that worked. The failure mode is not spending it, it is never rebuilding it, because the emergency is over and the urgency went with it. The practical fix is to decide in advance that refilling goes to the front of the queue the month after the money is spent, ahead of the goal payment and ahead of anything optional. On the practice file that means the same $1,022.84 that was building the floor turns around and rebuilds it, which is the whole reason the number is worth automating rather than remembering.
What comes after the number
The full method, including the shock pricing, the scoring sheet and the automation, runs across three chapters of Paycheck to Plan on one household followed the whole way through. Every figure in it was computed in code rather than typed, which is why the columns re-add.
If you would rather not buy a book, the seven-line subtraction above is the part that actually changes things, and it is free. Do that one first.
Questions people actually ask
Is an emergency fund three to six months of salary?
No. The common phrase is three to six months, but of a stripped-down survival month rather than of income. On the practice household here, the survival month is $4,000.00 while the ordinary month is $6,146.84. Sizing off income would have produced a target roughly half as big again, and a target nobody ever reaches is not a plan.
What should a first emergency fund target be?
One common method is to take the larger of two numbers: one bare survival month, and the single largest emergency you can realistically price. On the practice file that is $4,000.00 against $3,600.00, so the starter target is $4,000.00. That is a method, not a recommendation about your money.
Where should emergency money be kept?
The usual test is three-part: it should be safe, it should earn something rather than nothing, and it should be one or two days away rather than instant. Fast enough for a real emergency, slow enough to stop an impulse. Which specific account meets that test is a question for you and, if the details matter, a licensed professional. This page names no bank and no product.
Should I pay off debt or build savings first?
That depends on facts a page cannot see, including rates, stability of income and what a missed payment would trigger. What the arithmetic can show is why many households build a small starter cushion first: without one, the next unexpected bill goes straight back onto the debt that was being paid off. Run both orders on your own numbers before deciding, and talk to a nonprofit credit counselor if the debt is unmanageable.
Is this financial advice?
No. This is general financial education. It is not financial advice, not tax advice or investment advice, and not a recommendation about your situation. Every dollar figure is an invented practice number for a made-up household.