A sinking fund is money you set aside a little at a time for something you already know is coming. The car insurance renewal in March. The vet’s annual visit. The laptop that will not survive another year. The expense is not a surprise. What surprises people is the size of it landing in a single month.
The name comes from corporate finance, where a company sets money aside over years to repay a bond at maturity. The household version is the same mechanic at a smaller scale: a known amount, a known date, and a monthly transfer sized so the money is there when the date arrives.
Sinking fund versus emergency fund
These get confused constantly, and treating them as one pot is how emergency funds get drained by things that were never emergencies.
An emergency fund covers what you cannot forecast. A redundancy, a burst pipe, an urgent medical bill. Its size is a rule of thumb rather than a calculation, because you are insuring against an unknown.
A sinking fund covers what you can forecast but cannot absorb in one pay cycle. You know the amount, or close enough. You know the month. There is no uncertainty to insure against, only a timing mismatch between when the money arrives and when it leaves.
The practical difference shows up when the bill lands. Paying an annual premium out of an emergency fund feels like the same transaction, but it leaves you exposed until you rebuild the fund. Paying it out of a sinking fund leaves the emergency fund untouched, which is the entire point of having built one.
Which expenses actually deserve one
The test is simple: does it recur or is it certain, and is it big enough that one month’s income would not cover it comfortably?
Common ones that clear the bar are annual or semi-annual insurance premiums, vehicle registration and servicing, property taxes where they are not escrowed, holiday and Christmas spending, a car replacement, home maintenance, school costs, and the annual renewals of subscriptions billed yearly rather than monthly.
Things that fail the test are anything genuinely unpredictable, which belongs in the emergency fund, and anything small enough to absorb in the month it happens, which belongs in ordinary spending. A sinking fund for a $40 annual fee is bookkeeping theatre.
The categories people miss most often are the slow ones. Replacing a car and replacing a roof are both certain, both expensive, and both far enough away that no monthly budget line ever gets created for them. Those are exactly the expenses that end up on a credit card at a rate no savings account can offset.
Where the money should actually sit
This is where most sinking fund advice stops at “open a separate savings account,” which quietly costs people real money.
The FDIC publishes what American deposits actually pay. In its August 2026 table, dated 17 August 2026, the national average savings rate was 0.38%. Interest checking averaged 0.07%. Money market accounts averaged 0.63%. In the same table, the benchmark yield the FDIC applies to those non-maturity deposits, which its own footnote identifies as the effective federal funds rate, was 3.63%.
That gap is the whole decision. On a $10,000 sinking fund, 0.38% is $38 over a year. At 3.63% the same balance would produce $363. Nothing about the money changed, only which institution is holding it.
Two things follow. First, a sinking fund does not belong in checking, where the national average is 0.07% and the balance is one impulse away from being spent. Second, the “separate savings account” at the same big bank that holds your current account is usually the 0.38% account. Moving the money to an online bank’s high-yield account or a named high-yield savings product costs one afternoon and is the single highest-return action available on a sinking fund. Money market accounts are worth comparing at the same time, since the national average is already higher than plain savings and some come with check access.
Outside the US the same test applies with a different benchmark. The Bank of England’s Bank Rate was held at 3.75% at its most recent decision, so a UK saver whose instant-access account pays a fraction of that is making the identical mistake in sterling.
One account or one per goal
The internet loves the screenshot of eleven named savings accounts, one per goal. It looks organised. It mostly is not.
Every extra account is another login, another set of terms, another minimum balance to watch, and another statement at tax time. What it does not do is change how much interest the money earns or how much of it there is. A single high-yield account plus a short ledger of what each portion is earmarked for gives the same information with a tenth of the friction, and it is the natural extension of giving every dollar a job under zero-based budgeting.
The exception is worth stating clearly. Separate accounts earn their keep when a goal is far enough out and fixed enough in date to justify a different product, or when a shared goal needs a joint account that a partner can see and pay into.
One thing to avoid entirely: leaving a sinking fund as a balance inside a payment app. Stored balances are not the same as a bank deposit, the interest is usually nothing, and the protection question is different from the one that applies to a bank, which is worth understanding before you park months of savings there. The safety questions around Cash App balances apply to every app balance, and they are covered across the banking basics guides.
When a CD is the right home, and when it is not
Fixed-term deposits pay more at some maturities. The FDIC’s August 2026 averages were 1.41% for a six-month CD and 1.71% for twelve months, against 0.38% for plain savings, and competitive rates run well above those national averages.
The catch is the penalty. A sinking fund’s date is known, but “known” is doing a lot of work when the roof might fail early or the car might be written off. Lock the money only when the expense genuinely cannot arrive sooner than the maturity date: a school fee due in September, a tax bill with a statutory deadline, a planned wedding.
For goals more than a year out with a hard date, comparing current CD rates is worth the twenty minutes, and longer high-yield CD terms can beat instant access by enough to matter on a five-figure balance. For anything that might be needed early, the savings account wins on flexibility even at a lower headline rate.
The protection limits nobody checks
Sinking funds grow quietly, and people spread them across banks chasing rates without checking where the guarantee stops.
In the US, FDIC deposit insurance covers deposits automatically to at least $250,000 at each insured bank. In the UK, the FSCS automatically compensates up to £120,000 per eligible person, per bank, building society or credit union, and that limit applies to the total across all your accounts within the same banking group rather than to each account. In Australia, the Financial Claims Scheme protects deposits up to $250,000 for each account holder at each bank, building society and credit union.
For most sinking funds these ceilings are irrelevant. They stop being irrelevant when a house deposit is sitting alongside them, and the trap in both the UK and Australia is the same one: two brands sharing a single banking licence share a single limit.
Making it actually run
Size the monthly transfer by dividing the expected cost by the months until it is due, then round up. Automate it for payday rather than month end. Review the amounts once a year, because premiums and prices move and a fund sized in 2024 will be short in 2026.
Write down what each portion is for, in one place, in whatever tool you already open. The ledger is what turns a savings balance into a sinking fund. Without it you have a pile of money and a vague intention, and vague intentions get spent.
