Open any personal finance guide and the first instruction is the same: build an emergency fund of three to six months of living expenses, in cash, before doing anything else.

It's stated with such uniformity that it reads as settled fact. It isn't, and for a lot of households the standard version is either unachievable or the wrong thing to do first.

Where the number came from

Nowhere in particular, as far as anyone can establish. There's no study demonstrating that three to six months is optimal. It appears to be a reasonable-sounding rule of thumb that got repeated until it became doctrine.

The logic behind it is sound enough — the most common financial shocks are job loss and large unexpected costs, and a few months of runway covers a lot of both. But the specific figure is a convention, not a finding.

The problem is that a convention presented as a requirement discourages the people who need help most. Somebody living close to the edge, told they need six months of expenses saved, quite reasonably concludes the whole framework isn't for them and disengages.

What the research actually suggests

The more interesting work in this area looks at what level of savings actually reduces financial distress, and the findings are encouraging.

The largest benefit comes from the first tranche. Having anything — a few hundred, a thousand — dramatically reduces the probability of a small shock cascading into a serious problem, because it's the difference between paying for a car repair and taking on high-interest debt to pay for a car repair.

Going from nothing to a modest buffer changes outcomes substantially. Going from three months to six months changes them much less.

Which suggests a completely different framing: get to a small buffer as fast as possible, then reassess, rather than treating six months as a gate you have to pass before anything else happens.

The debt question

The place where the standard advice most clearly fails.

If you have credit card debt at a high interest rate, holding a large cash reserve at near-zero interest is a guaranteed loss. Every month, the balance costs you more than the savings earn.

The rational sequence is: build a small buffer, perhaps a month of essential expenses, then attack high-interest debt aggressively, then build the fuller reserve.

The counter-argument is behavioural, and it's not trivial. Without a buffer, the next unexpected cost goes straight back onto the card, and you're in a cycle. That's why the small initial buffer matters — it breaks the cycle without holding an expensive amount of idle cash.

The employer pension question

Another common failure of the strict sequence. If your employer matches pension contributions, declining that match to build savings faster is giving up a guaranteed immediate return that no savings account can approach.

In most circumstances, contributing enough to capture the full match should come before building a large emergency fund. The money is illiquid, which is the trade-off, but the size of the return is difficult to argue with.

Who needs more, who needs less

The other flaw in a universal figure is that risk varies enormously between households.

You probably need more than six months if: your income is variable or commission-based; you're self-employed; you work in a sector with long re-employment times; you're a single earner supporting others; you have significant health issues; or you own a home with ageing systems.

You probably need less if: you have very stable employment with strong statutory protection; you have two incomes in unrelated sectors; you have low fixed commitments and could reduce spending quickly; you have access to family support you'd actually use; or you rent and could move somewhere cheaper.

That second list describes a lot of younger people, and telling them to sit on six months of cash before investing anything can cost them a significant amount over a long horizon.

Where to keep it

The practical part, and one where a lot of people leave money on the table.

An emergency fund needs to be accessible within days and protected from loss. It does not need to be in a current account earning nothing.

Instant-access savings accounts, money market funds and similar instruments provide meaningful returns with same-day or next-day access. The gap between a savings rate and nothing, on several thousand, is a real amount of money for zero additional risk.

What it shouldn't be in: anything that can fall in value, anything with a withdrawal penalty, or anything you'd feel emotionally reluctant to sell. The point of the fund is that using it is uncomplicated.

A more useful framing

Rather than a fixed multiple, I'd suggest thinking in terms of the specific shocks you're actually exposed to.

What's the largest plausible unexpected cost in your life — a car, a boiler, a deposit on a new flat if you had to move? Cover that first. Then think about income interruption: how long would it realistically take you to find comparable work, and what would you actually spend during that period, given you'd cut discretionary spending immediately?

That number is usually lower than six months of current expenses, because current expenses include things you'd stop paying for. And it's specific to you, which is more useful than a rule that was never derived from anything.