Sinking funds: how to stop being surprised by insurance

Ask yourself what you spent on gifts last year. Not in any particular month — the total.

Most people can’t name the order of magnitude. Yet almost everyone reaches several hundred dollars: birthdays, the holidays, weddings, friends’ children. The sum was simply never assembled in one place, because it left in fragments.

The same goes for insurance, tyres, the dentist, the vet, and replacing whatever broke. Each of those looks like an emergency inside a single month. Across a year they form a pattern, and a fairly predictable one.

Why they wreck a month

The size isn’t the issue. The distribution is.

An ordinary month contains none of them. You get used to that, and your sense of what normal spending looks like forms from months with no irregular events in them.

Then a month arrives where insurance and the dentist coincide. Spending doubles, the budget falls apart, and you conclude that budgets don’t work. What didn’t work was your idea of a normal month: it was built from a sample with every such month excluded.

Unusual months happen roughly one in three. That isn’t an exception to the rule. It’s the part of the rule you never counted.

Calculating your number

Half an hour and a year of statements.

Go through twelve months and write down everything that doesn’t recur monthly. By type: insurance of any kind, car servicing and repairs, medical beyond the routine, the vet, gifts, travel, electronics, significant clothing, home repairs, courses.

Add it up. Divide by twelve. That figure is your monthly contribution.

Illustratively: insurance $520, tyres and servicing $680, gifts $840, dental $610, laptop repair $420. That’s $3,070 over the year, or roughly $256 a month.

The first reaction is usually the same: $256 a month sounds impossible. But you’re already spending it. The only difference is that right now it arrives in lumps and destroys a specific month each time.

With no year of history, start from a rough estimate and refine after three months. An approximate figure works; zero doesn’t.

Where to keep it

Separate from everyday money. That’s the one non-negotiable requirement.

If the fund sits on the card you buy groceries with, it will get spent. Not deliberately, but gradually: the balance looks large and purchases look affordable.

Workable options: a separate savings account, a separate card, an accessible deposit. The yield doesn’t matter. Two properties do — the money is visible separately, and you can retrieve it within a day when the event arrives.

What not to do: keep it somewhere it can’t be withdrawn quickly without penalty. The point of the fund is readiness, not return.

One fund or several

An argument settled by temperament rather than theory.

One combined fund is simpler. One amount, one account, one contribution. The cost is that a large event empties it entirely, and the next event meets nothing.

Several targeted funds are clearer: car separately, medical separately, gifts separately. An event only drains its own fund. The cost is administration — three or four accounts, or manual tracking of shares inside one.

The sensible compromise for most people: one account with the division kept on paper. The money lives in one place while a note records how much of the total belongs to the car and how much to medical. Same logic as virtual envelopes, covered in envelope budgeting.

How this differs from an emergency fund

People conflate them constantly, and the functions are different.

An emergency fund covers lost income. It goes untouched for years and rebuilds slowly.

A sinking fund is spent regularly and is supposed to be. It isn’t savings, it’s deferred spending: the money is already assigned to specific events that simply haven’t happened yet.

Keeping them together is awkward for exactly that reason. You’ll count your emergency fund as larger than it is, then discover at the moment of genuine need that half of it was promised to winter tyres.

The order of building them differs too. First a small emergency fund covering one month of fixed costs, then the sinking fund, and only then growing the emergency fund further.

Getting through the first year

While the fund fills, events will arrive before the money does. That’s normal and doesn’t mean the approach failed.

The practical route: start with the most predictable event. If insurance renews in March at $520, save toward that with a known deadline. One event covered end to end builds more confidence than an abstract fund ever will.

Then add one line at a time. Within a year you’ll have the full set, and months will stop dividing into “normal” and “the ones where everything broke.”

There’s a useful side effect. Assembling annual totals, many people see the real scale of certain categories for the first time. $840 a year on gifts isn’t a reason to stop giving gifts; it’s a reason to decide whether that figure suits you. Making decisions like that from data is the subject of cutting wasteful spending.

Running it in the app

Voice Finance has no sinking-fund feature, and this approach doesn’t require one.

Don’t record the contribution as an expense. It’s a transfer between your own money, and logging it as spending inflates the month twice. The expense appears later, when the event arrives and the money genuinely leaves.

At that point, record an ordinary transaction with its real category: insurance, medical, gifts. The month’s statistics then tell the truth, and a year later the annual breakdown in Analytics gives you accurate figures for recalculating the contribution.

The fund’s balance itself is easier to watch in your banking app, where the money actually is, rather than duplicated in expense records. Folding the contribution into the wider calculation is covered in building a monthly budget.

Start with one line

Don’t total the whole year today. Take the nearest known event: insurance, a service, someone’s birthday in three months.

Divide the amount by the months remaining and start setting it aside. When the event arrives and the money is already there, you’ll understand the mechanism better than any calculation could explain it, and the remaining lines will follow on their own.