How Big Is My Emergency Fund? The Calculator, the Formula, and the Months to Target — Parking Rate Dated October 2026

A repair estimate of $1,400 lands on a Tuesday, and the month was already balanced on the assumption that nothing would break. “How big should my emergency fund be?” is the question that arrives with the bill — and most answers to it are moods dressed as advice: some months of expenses, a rough cushion, enough to feel safe. You cannot build toward a mood. The reserve is built from a number, not a mood, and this page is that number as arithmetic: the inputs, the formula shown, a working calculator, and the dated 2026 rate the reserve parks at while it is being built.

Three questions get asked over and over, so here they are before the math. How big is enough? Essential outgo — the money that leaves the house no matter what — multiplied by three to six. Why six months? Because six is the end of the range where the shock is large: the 2026 individual out-of-pocket maximum of $10,600 — the most a health plan can make you pay yourself in a year — a national vehicle fleet averaging 12.6 years, a paycheck that can go quiet. Why not six for everyone? Because three is the measured floor — the yardstick the Federal Reserve’s own household survey uses — and Citi’s May 29, 2026 guide says what the whole industry repeats: three months if your income is stable and your obligations are few, six or more if your income fluctuates or you have dependents. The choice between the ends of the range is the subject of How Big Should Your Emergency Fund Be? — this page is the calculator that takes the choice you made and turns it into months.

The Inputs

The calculator takes five numbers. That is the whole input set, and every one of them is yours, not a survey’s:

The default set loaded into the calculator below is not made up. It is the worked household from this site’s sizing article, rebuilt from its dated rows: $2,000 national average rent (Zillow, March 2026), $884 a month from the 2026 individual out-of-pocket maximum of $10,600 spread over the year, and a stated $900 planning line for food, utilities, transport and phone. That is $3,784 of essential outgo. The contribution default is $503 — 10 percent of $5,027, the Bureau of Labor Statistics’ Q2 2026 median monthly earnings for full-time workers aged 25 to 34. Swap your own rows in; the arithmetic does not care where your rent came from, only that you used yours.

The Formula, Shown

Four lines. If you can read them, you can run the calculator by hand on a napkin — and you can check any box that claims to be one.

essential outgo = the sum of your essential monthly lines
reserve target  = essential outgo x the multiple in months
gap             = reserve target − amount already saved
months to target = gap ÷ monthly contribution

Rounding, stated once: each monthly line and the contribution are whole dollars; the target is exact (a whole-dollar outgo times a whole number); months are shown to one decimal, and the transfer that crosses the line is a whole payday — 22.6 months means the 23rd transfer lands you there. That is the whole model. No hidden factor, no assumed market return, no smoothing.

The Calculator

The box is the formula with a text field on each input. If it renders as an empty frame in an old browser, nothing is lost that the printed arithmetic does not already carry — and the noscript line inside it does the worked example in ink.

The Worked Output

One set of inputs — $3,784 essential outgo (the dated worked set), $0 saved, $503 a month — and the two ends of the range. Every output recomputes from the four lines; you can check each one in a minute.

Read the table for what it says about waiting, because this is what the moods leave out. The build is linear: at a fixed transfer, every month you do not start is a month the target moves back by exactly one month. There is no version of “I’ll start when things calm down” that does not finish in the same month — one year further out. The shock does not consult the schedule. That is the whole argument for making the transfer automatic on payday, before the discretionary money exists to spend: the cushion first, the clever stuff after.

And once the reserve lands, the transfer does not retire — it moves. What the clever stuff is, starting at the first $500, is the next article in this tree.

The Parking Rate: The Cost of the Wrong Address, Dated October 4, 2026

The target and the months are the reserve’s size. The parking rate is where the reserve actually lives, and it is the input this page re-dates. From Bankrate’s weekly survey of more than 500 banks and credit unions, as of October 4, 2026: the national average savings yield is 0.66% APY, and the best high-yield offers on the table pay around 4%. The same survey read 0.64% on September 28, 2026 — that is what a weekly survey looks like when it is honest: it moves when the Federal Reserve moves, and the Fed sits at a 3.75–4.0% range as of September 16, 2026, the date on this site’s rate set. The table re-dates with the Fed, not before.

The gap between those two lines is not an abstraction. It is the fifth input run at its two ends:

• Parked at the 0.66% average, the $22,704 six-month target earns about $150 a year. At ~4%, it earns about $908. The difference — about $758 a year — is what the wrong address costs on the full shield. At the 3-month floor, about $379 a year.
• Now put the same lines next to the shock. A $3,000 repair carried on the 24% card runs at about $720 a year of interest; held against a reserve earning ~4%, the same $3,000 runs at about $600 a year of spread — and at the end of the year the reserve is still yours, while the card keeps the $720. That is the 24% card’s whole trick: always open, priced at twenty points above a ~4% account and more than twenty-three above the 0.66% one.
• The honest other side, because a rate is a rate: parked at ~4%, the full shield gives up roughly $454 a year against the 6% real long-run record this site’s dated set uses for the market — about $227 at the 3-month floor. (The sizing article’s figure of about $1,360 counts the same trade before crediting the reserve its own ~4%; credit the parking and $454 is what is left. Same arithmetic, one step further.) That is the price of the shield, counted rather than hidden. The reserve is not money you are trying to grow; it is money you are trying not to lose.

While the reserve fills, the parked money earns too. At ~4%, the three-month floor collects about $396 of interest during its own 22 months of construction — the transfer builds the reserve, and the reserve shortens the build.

The Specification This Page Sets

This article is the calculator’s specification, in the open, so the box and the printed math can never quietly disagree. The box above runs exactly the four lines, and every calculator page this site publishes runs them the same way: five inputs (outgo, saved, contribution, multiple, parking rate); whole-dollar rounding of the monthly lines and the contribution; an exact whole-dollar target; months to one decimal plus the whole payday that crosses the line; a completion month projected from the current date; interest credited at the stated rate shown as its own line; and every rate on the page carrying the date of the weekly Bankrate survey, re-dated when the Federal Reserve moves. A calculator that cannot be reproduced from four printed lines is not a calculator — it is a mood with a text box.

Check It Against the Dated Set

The sizing lives in How Big Should Your Emergency Fund Be? — the 3-versus-6 disagreement, the dated rows behind $3,784, and the months table this calculator reproduces at its defaults. And when the reserve lands, the cushion is what lets the investing start: the first $500.

The reserve is not a reward for being good at money. It is the thing that lets you actually do the rest. Run the four lines with your own rows — the number to build it, and the account to park it in, rate dated. The cushion first, the clever stuff after.