$1,000 buffer first, or every dollar at the card? The insurance premium is $346

Same $4,100 of card debt, same $230 a month. One question: before you throw everything at the card, should you park $100 a month into a small buffer first? Here is what the arithmetic actually says — no advice, just the numbers, so you can run your own.

The setup

• One card: $4,100 at 18.9% APR
• Monthly budget: $230, fixed, no new charges
• Buffer plan: $130 a month at the card plus $100 a month into a buffer, for 10 months, until the buffer hits $1,000 — then the full $230 goes at the card

Assumptions, stated so you can check my math: interest accrues monthly at APR / 12. Same model as the earlier pieces in this series, so the baseline is the one already published: $230 a month straight at the card clears $4,100 in 22 months, $751 of interest, $4,851 total.

Path A: every dollar at the card

• Debt-free in 22 months.
• Total interest: $751. Total paid: $4,851.
• Cash buffer at the end: $0.

Path B: build the $1,000 buffer first

• The buffer reaches $1,000 in month 10.
• Debt-free in 27 months.
• Total interest: $1,097. Total paid to the card: $5,197.
• Cash buffer at the end: $1,000.

The buffer costs $346 of extra interest and 5 extra months. That is the premium on the insurance. You pay it whether or not anything goes wrong.

What if something goes wrong?

An $800 car repair is the classic test. Two timings:

It hits in month 4. Path A has no buffer, so the $800 goes back on the card: 26 months, $1,043 of interest. Path B has only $300 saved by then — $500 still goes on the card: 31 months, $1,337 of interest. A half-built buffer does not cover an $800 emergency. The buffer only helps once it is actually big enough.

It hits in month 12. Path A again puts the $800 on the card: 26 months, $911 of interest. Path B pays it from the $1,000 buffer and the card never sees it: 27 months, $1,097 of interest, $200 left in the buffer. Even here, the buffer path still pays more interest overall ($1,097 vs $911) — because the slower payoff accrues interest the whole way. The buffer did not save money. It saved you from new debt.

That is the honest shape of it. The buffer is not a money-saving move. It is a liquidity move: cash in hand when the repair shows up, instead of the card balance going back up. The price of that liquidity, on these numbers, is $346 and 5 months — if nothing goes wrong. If something goes wrong early, the price is higher, because the buffer is still too small to do its job.

Sizing it

One reader made a good point on an earlier piece in this series: measure a buffer in months of essential spending, not in round numbers. $1,000 here is just a round number I picked for the arithmetic. The sizing logic is theirs and it is the right way to think about it — and the arithmetic scales the same way whatever size you pick: every dollar parked in the buffer is a dollar not cutting interest, so the premium grows with the buffer.

The difference

If you are certain nothing will go wrong for two years, the buffer is $346 wasted. If you are not certain — and a half-built buffer is worse than none at all in the early months — the question is whether $346 is a fair price for not putting the next emergency back on the card.

Run your own numbers

I ran these on the same model behind a free payoff calculator I built: https://alapha888.github.io/irregular-income-debt-baseline/ — enter your balances, APRs, and monthly payment and it shows the months and interest. No signup, no email. The calculator has no buffer switch, so the buffer numbers above are the same model run by hand with the payment split.

The model is simple on purpose. Real cards vary — minimum-payment formulas, penalty APRs, promotional rates — so treat this as an illustration, not a plan. If the numbers change your mind about anything, that is the point: the arithmetic is checkable, and it is yours to redo.