How much should you keep for a rainy day?

Every financial plan eventually meets a week it didn’t plan for. The transmission goes. A child needs stitches. A company announces “restructuring” on a Tuesday. None of these are unusual; over a decade, most households see several. What varies is whether they arrive as an inconvenience or a crisis, and the difference is almost always a pile of cash that was sitting there, doing nothing, on purpose.

That pile is an emergency fund. It earns less than your investments, it feels idle, and it’s one of the few pieces of personal finance advice that nearly everyone, from Dave Ramsey to the Bogleheads to your grandmother, agrees on.

Why it matters more than it looks

Buffett was talking about companies, but households work the same way. In good times, it’s hard to tell a family with a cushion from one without. The difference shows up only when something goes wrong.

The Federal Reserve asks Americans every year how they’d cover an unexpected $400 expense. In its 2025 survey, 63% said they’d pay it with cash or its equivalent, the same as the three years before. The rest would borrow, sell something, or couldn’t cover it at all. A $400 surprise is not a disaster. Put it on a card at 24% because there’s no alternative, though, and it becomes one slowly.

An emergency fund also does something less obvious: it protects your other plans. Without one, every surprise gets paid for by raiding the vacation fund, pausing retirement contributions or running up a card, and the plan you worked hard on quietly unravels. With one, the surprise is absorbed, the plan carries on, and you refill the fund over the next few months.

There’s a psychological payoff too. Morgan Housel argues in The Psychology of Money that the highest dividend money pays is control over your time: the ability to say no to a bad job, wait out a bad market, or take a month to find the right next step. An emergency fund is the first, smallest version of that. It’s a few months in which you don’t have to make a desperate decision.

How much: three months, six, or more?

The standard advice is three to six months of expenses. That range is fine as far as it goes, but it hides two important details.

First, it means essential expenses, not total spending. If you lost your income, you’d stop eating out, pause the gym and skip the trip. What you couldn’t stop is rent or mortgage, utilities, insurance, groceries, childcare, minimum debt payments, transportation and medical costs. Total those for a normal month; that’s the unit. For many households it’s 60 to 75% of what they actually spend.

Second, where you land in the range should depend on how likely you are to need it, and for how long. The average spell of unemployment in the US has run around five months in recent years, and longer for senior roles, which can take a while to replace at the same pay. A few questions will tell you which end of the range is yours:

• How many incomes? Two incomes from different employers rarely disappear at once. One income means no backup, which argues for six months or more.
• How stable is the income? A tenured teacher and a commission-only salesperson need very different cushions. Freelancers and business owners often keep six to twelve months.
• Who depends on you? Children, a partner at home or an aging parent raise the stakes of any gap.
• How specialized is your job? The longer it would take to find an equivalent role, the bigger the fund should be.
• What else could go wrong? An older car, an older house, a health condition, or a high insurance deductible all raise the odds of a large surprise.
• Is a baby on the way? Unpaid or partly paid leave is a planned income gap, and it’s better funded from a separate sinking fund so the emergency fund stays intact.

Some people go further. Suze Orman now recommends at least eight months of living expenses, and up to twelve, a target she raised after watching how long people were out of work in 2020. That can feel excessive in a strong job market; it rarely feels excessive in a weak one.

What counts as an emergency

A useful test has three parts. It’s unexpected, it’s necessary, and it’s urgent. A furnace failing in January passes all three. Holiday gifts fail the first: December arrives every year. A new TV fails the second. A great deal on flights fails the third.

Most “emergencies” that drain funds are really irregular expenses that were never planned for: car registration, the annual insurance premium, the dentist. These deserve their own sinking funds, set aside month by month. Keeping the two separate is what lets the emergency fund actually be there for emergencies.

Job loss is the big one, and the reason for the months-of-expenses sizing. Medical bills, urgent home and car repairs, emergency travel for family, and a gap between jobs round out the list.

Where to keep it

An emergency fund has one job: be there, in full, when you need it. That rules out anything that can fall in value at the wrong moment, which is why it doesn’t belong in the stock market. The bad month for your job is often a bad month for stocks too. It also shouldn’t sit in your everyday checking account, where it blurs into spending money and quietly gets spent.

• A high-yield savings account at a separate bank is the default for good reason. It’s FDIC-insured up to $250,000 per depositor, per bank, per ownership category, it pays far more than a typical big-bank savings account, and money can usually reach checking in a day or two. The slight friction of a different bank is a feature.
• Money market funds and Treasury bills can pay a little more and suit larger funds. They aren’t FDIC-insured, but government money market funds and T-bills are about as safe as money gets outside a bank.
• Some people keep part of a large fund in a tiered setup: a month or two in savings, the rest in T-bills or a money market fund.
• A Roth IRA is sometimes suggested as a backup, because contributions (not earnings) can be withdrawn without tax or penalty. It works, but money pulled out can’t easily be put back, so treat it as a last line, not the plan.

Whatever you choose, give the account a name like “Emergency fund, not for spending.” It sounds silly and it works.

How to build it without stalling everything else

Six months of essentials is a big number, and saving it can feel endless. It helps to do it in stages rather than all at once:

1. 1

   Start with a starter cushion

   Aim for $1,000 to $2,000, or one month of essentials. This alone handles most car repairs and medical bills, and it breaks the habit of reaching for a credit card.

2. 2

   Keep the employer match going

   If your employer matches retirement contributions, keep contributing enough to get the full match while you build the fund. It’s part of your pay.

3. 3

   Clear high-interest debt

   Paying off a card at 20% or more is a guaranteed return few investments offer. With a starter cushion in place, you’re less likely to run it back up.

4. 4

   Grow to the full target

   Then build to three to six months or more, depending on your answers above.

5. 5

   Invest the rest

   Once the fund is full, money that was going into it can go to long-term investing and other goals.

Automate the transfer for the day after payday, so the money is gone before you can miss it. Send windfalls straight to the fund: tax refunds, bonuses, gifts, the money from selling the old stroller. A $3,000 tax refund can cover months of transfers in one go.

And give it a clear finish line. “Save more” is not a goal. “$18,000 by next June” is, and it tells you exactly what each month needs to be.

The case against a big fund, and why it mostly fails

The best argument against a large emergency fund is opportunity cost. Over long periods, stocks have beaten cash by a wide margin, so $30,000 in savings could have grown far more invested. Some people say they’ll use credit cards or a home equity line in an emergency instead.

The math on expected returns is right; the plan usually isn’t. Lenders cut credit lines in recessions, exactly when you’d need them, and borrowing when you’ve lost your income is a hard road.

The same downturn that costs you a job often knocks 30% off your investments, so you’d be selling low. And the benefit of a fund is not only financial: it’s the ability to make good decisions calmly. That’s hard to price, but it’s real.

After you use it

Using the fund is not a failure; it’s the fund doing its job. The only mistake is not refilling it.

Once the emergency passes, pause the less urgent goals for a few months and point the money back at the fund until it’s whole.

Then take a moment to ask whether the thing that happened is likely to happen again. If the car needed $2,400 of repairs this year, it may be time for a car-replacement sinking fund too.

Revisit the target once a year, and after any big change: a new baby, a move, a raise, a new job, a partner leaving work. The right size for your fund is not fixed. It moves with your life.

Keeping an eye on it

In Foundation, an emergency fund is one of the goal types. Link it to the savings account that holds the money and it shows progress against your target as the balance changes, while keeping that account out of your safe-to-spend figure so it never looks like spending money. Foundation can’t move money, so the automatic transfer is set up at your bank.