How Big Should Your Emergency Fund Actually Be?

The standard advice is three to six months of expenses. It is good advice, given badly, because the word “expenses” does most of the damage.

Essentials, not outgoings

Calculate the target from your total monthly spending and you get a number large enough to feel hopeless. It is also the wrong number.

An emergency fund covers what continues when income stops. In that situation the subscriptions go, the restaurant meals go, the gym membership goes, the holiday is cancelled and the clothes budget disappears.

What remains:

• Rent or mortgage
• Utilities and council tax
• Food and household basics
• Transport, including getting to interviews
• Insurance
• Minimum debt payments, so your credit survives
• Childcare, if stopping it would prevent you working

For most households, essentials come to 55% to 75% of total spending. Working from the right base often cuts the target by a third, which is frequently the difference between a goal someone starts and one they look at once and abandon.

Three months or six, and when it should be more

The range exists because the right answer depends on how fast your income could stop and how long it would take to replace.

• Two stable salaries, no dependants: three months is defensible. Both incomes rarely stop at once.
• One salary: four to six months. There is no second income underneath you.
• Contract, commission or seasonal work: six months. The income is lumpy and the gaps are the entire point of the fund.
• Self-employed: six to twelve months. No notice period, no redundancy payment, often no sick pay.
• Dependants, a specialised role, or a large mortgage relative to income: add to whichever of the above applies. Specialised roles pay well and take longer to replace, which cuts both ways.

There is an upper bound worth respecting. Beyond about twelve months of essentials, cash becomes a poor place for money and the inflation cost starts to outweigh the security.

The order of operations

Most people are told either “build the fund first” or “kill the debt first”. Neither works on its own, and the sequence that does is a compromise.

1. Take any employer pension match. A 100% match is an instant doubling. Nothing else on this list beats it, including a 26% credit card.

2. Build a starter buffer of about one month of essentials. Roughly 1,000 to 2,000 for most households. This absorbs the ordinary shocks: a car repair, a boiler, a vet bill, a broken laptop.

3. Clear high-interest debt. Anything above about 8%. Paying off a 22% card is a guaranteed 22% return, which no savings account will ever match.

4. Finish the fund. Back to three to six months, or whatever your situation calls for.

Step two is the one people skip, and skipping it is why “debt first” so often fails. Without a buffer, the next unexpected expense goes straight back on the card, and a month of repayment progress evaporates. That is not a discipline failure. It is a plan with no shock absorber, resetting on contact with an entirely normal event.

Where to keep it

Instant-access savings, at the best rate you can find, at a different institution from your current account.

Each part of that matters.

Instant access, because the whole value is availability within a day. A notice account or a fixed-term bond fails the one job the money has.

Best rate you can find, because instant-access rates vary enormously and the worst offenders are usually the bank people have been with for fifteen years. Moving 10,000 from 0.5% to 4.5% is 400 a year with no additional risk and no loss of access. This is close to free money and it takes an afternoon.

Different institution, because a balance you see every time you check your current account is a balance you will eventually spend.

Not invested. This is the one people argue about, and the argument loses on timing. Emergencies correlate with recessions: redundancies cluster exactly when markets are down. A fund that might be worth 30% less at the moment you need it is not performing its function.

What counts as an emergency

Two tests, and both must be true: is it unexpected, and is it necessary?

A boiler failing in January is both. A car repair you need to get to work is both. An urgent dental cost is both. Redundancy is obviously both.

A wedding you have been invited to for a year is expected. A holiday is not necessary. A new phone because the old one is irritating is neither.

These are things to save for separately. Using the emergency fund for them is the single most common way a buffer quietly ceases to exist, one reasonable-seeming exception at a time.

If you do use it legitimately, refilling it becomes the next priority. Not eventually. Immediately, ahead of other saving.

Does inflation not eat it?

Yes, and that is the cost of the insurance.

A buffer earning 4% while inflation runs at 3% roughly holds its value. One earning 0.5% loses about 2.5% a year in real terms. On a 10,000 fund that is 250 a year of purchasing power, which is a genuine cost and a completely reasonable price for what it buys.

Getting a competitive rate turns most of that loss into a rounding error, which is why the account you choose matters more than the theory does.

What you are actually buying

The financial return is poor by design. The value is elsewhere, and it is larger than it looks.

It removes the cost of borrowing in a crisis, which is exactly when borrowing is most expensive and hardest to arrange. It turns an unexpected bill into an inconvenience rather than a spiral through a credit card into a personal loan.

And it buys the ability to say no. To leave a job that is damaging you. To refuse work on terms you should refuse. To take a week to find the right answer rather than accepting the first one because rent is due.

That optionality is worth far more than the interest forgone, and it is why the buffer comes before nearly everything else.