Money basics

Sinking Funds: The Trick That Kills Surprise Bills

Almost no large expense is actually a surprise. How to name it, divide it by the months until it lands, and stop raiding your emergency fund every spring.

By the Euphoria team · 2026-07-26 · 8 min read

The list that was never a surprise

Every year the Federal Reserve asks American adults a plain question: could you cover an unexpected $400 expense using cash or its equivalent? In the 2025 survey, 63 percent said yes, a share that has not moved in four years. The other 37 percent said they would have to borrow, sell something, or simply not pay.

The word doing the damage in that question is "unexpected."

Write out the actual list of things that wreck a month and read it slowly. Car insurance, billed every six months. Vehicle registration and the state inspection. Tires, which wear out on a schedule you could look up. A dentist visit your plan does not fully cover. Holiday gifts. The laptop that will not last forever. A wedding you were told about in March for a date in August. Not one of those is unexpected. Every single one is either on a calendar or on a clock.

They arrive feeling like emergencies for one reason. Nobody divided them by twelve.

• 63% of US adults could cover a $400 emergency expense with cash or its equivalent in 2025
• 37% could not, a share that has not shifted in four years
• $303 what the six predictable expenses below cost per month once divided

Where the name comes from

A sinking fund is money you set aside on a schedule for a specific expense you already know is coming.

The phrase is borrowed from corporate finance, and the original is worth knowing because it makes the household version stick. A company that issues a bond owes the full face value back on a single day, years in the future. A company that waits for that day to go find the cash is a company in trouble, and lenders know it. So bond agreements often require the issuer to pay into a sinking fund along the way, retiring the obligation in pieces, so the maturity date arrives as a paperwork exercise rather than a crisis.

Your car insurance premium is a small bond that matures every six months. Your holiday spending is a bond that matures in December, every December, with no exceptions ever granted. The corporate version of you would have started funding it in January.

Doing the arithmetic once

A sinking fund needs exactly three numbers per expense: what it costs, how often it lands, and therefore what it costs per month. Only that third number has to stay in your head afterward.

Here is a realistic set for one person. The dollar amounts are illustrative, but the division is the whole technique.

• Car insurance, $1,800 a year billed in two payments of $900. Divided by 12 months, that is $150 a month.
• Holiday gifts and travel, $600, landing in December. Divided by 12, that is $50 a month.
• Dental and vision costs your plan does not cover, about $360 a year. Divided by 12, that is $30 a month.
• A laptop replacement, $1,200, needed in about four years. Divided by 48, that is $25 a month.
• Tires, $840, replaced roughly every three years. Divided by 36, that is $23 a month.
• Registration, inspection, and the annual subscriptions that all renew at once, $300 a year. Divided by 12, that is $25 a month.

Add the right-hand column and six separate ambushes become one line item of about $303 a month. That number is not small, and seeing it is the uncomfortable half of the exercise. But you were already paying it. You were paying it in lumps, at the worst possible moments, sometimes with interest attached. The only thing that changed is who chose the timing.

Look closely at the laptop and the tires, because they are where people get the arithmetic wrong. Neither is an annual expense, so dividing by twelve would be the wrong move. You divide by the number of months until the thing actually has to be replaced. A $1,200 laptop four years out is a $25 a month problem. The same laptop needed next month is a $1,200 problem. Same object, completely different question, and the only variable that moved is time.

This is not your emergency fund

An emergency fund covers the genuinely unforeseeable: the job that ends, the trip to the emergency room, the transmission that had no business failing at 60,000 miles. It is a buffer against the unknown, and the honest thing about the unknown is that you cannot fund it in neat pieces, because you do not know its size or its date.

A sinking fund is the opposite on both counts. You know roughly the size and roughly the date. That is precisely why the two should not live in the same pile of money.

An emergency fund is for the things you could not have known. A sinking fund is for the things you did.

Mixing them has a predictable failure mode, and it runs on a yearly cycle. Insurance comes due in April. The money comes out of the emergency fund, because that is where the savings are. The fund never gets refilled, because nothing forces it to. By the time an actual emergency turns up in November the buffer is thin, so the emergency goes on a credit card, and the household concludes it is bad at saving. It was not bad at saving. It was running one account for two incompatible jobs.

What the raid actually costs

Put a number on it, because the number is smaller than people expect, and that is exactly why the habit survives.

Say the $900 half-year premium goes on a card at 21 percent instead of coming out of a fund, and you pay $150 a month against it, which is the same $150 the sinking fund would have been collecting anyway. It clears in seven months and costs about $959 in total, so roughly $59 of interest.

Fifty-nine dollars is not a catastrophe. The calendar is. That premium was a six-month premium, so the next one arrives in month six, while you are still paying off the last one. Do that twice and you are not behind by $59. You are permanently one premium behind, carrying a balance that never quite reaches zero and paying interest on it forever. The interest is the symptom. The overlap is the disease.

What a working fund looks like

A sinking fund is supposed to empty itself. That sounds obvious and it still trips people up, because a savings balance that drops looks like a mistake.

Follow the insurance fund through one year at $150 a month.

The balance climbs for six months, reaches $900 exactly when the premium is due, pays it, and starts over. Two peaks, two drawdowns, and at no point in the year did you have to find $900 you did not have. Graded on the size of the balance, this looks like twelve months of getting nowhere. Graded on the job it was hired for, it is perfect.

That distinction matters more than it sounds, because it is the reason people abandon the method in month seven. The fund hits zero, it feels like a setback, and the transfer gets canceled. Nothing went wrong. The fund did the only thing it was ever meant to do.

Where to keep it and how to move the money

Two arrangements work, and choosing between them is a question about your tolerance for admin rather than about the money.

The first is separate named accounts. Many banks and credit unions let you open several savings accounts, or sub-accounts inside one, at no cost and with your own labels, so you end up with a bucket called Insurance and a bucket called Tires. The advantage is structural: you cannot casually spend the tire money, because the tire money is visibly not in checking. The CFPB makes the same argument in its guide to building an emergency fund, which recommends keeping savings somewhere safe, accessible, and out of easy reach of everyday spending.

The second is one savings account plus a spreadsheet. One real balance, several imaginary columns that add up to it. Fewer accounts to open and fewer logins to manage, but it only works if you actually maintain the columns, because an unallocated balance is a balance you will eventually spend.

Either way, two mechanics do the real work. Automate the transfer, and schedule it for the day after payday. The CFPB suggests a recurring transfer you set up once and leave alone, and the timing is behavioral rather than financial. Money that leaves before you have started spending the paycheck never registers as a sacrifice. Money you have to remember to move on the 28th is competing with everything else you wanted that month, and it loses.

One more habit worth building in. When an expense finally lands and the fund drains, do not pause the transfer to enjoy the breathing room. The next premium started accruing the day the last one was paid.

What changes once you have done this

The useful output of this exercise is not the $303 figure. It is that the phrase "unexpected expense" gets a lot narrower.

Once six recurring costs are named, priced, and divided, the only things left in the genuinely-unforeseeable category are the things that actually belong there. That is a smaller and much more manageable list, and it is the list your emergency fund can realistically cover. Most households do not have a savings problem so much as a classification problem: they are treating scheduled costs as shocks, and then judging themselves for being surprised.

You can run this on your own numbers in about twenty minutes with a notes app. Inside Euphoria, the budgeting lessons let you build the same table against a simulated paycheck and watch what happens when you skip a month, which is a faster way to feel why the timing matters than finding out in April.