This question is the personal-finance equivalent of “should I run or do strength training” — the answer is usually “both, in a specific order, and the order depends on the specifics.”

Here are the specifics that matter.

The math, briefly

Carrying a balance on a 22% APR card costs you 22% per year on whatever you carry. Putting money in a high-yield savings account currently earns roughly 4% per year. So $1 in savings vs $1 paid off the card: the card win is 22% - 4% = 18% per year.

For pure expected-value, paying off the card wins by a wide margin. If that were the whole story, the answer would be “always pay off the card first.”

It isn’t the whole story.

The case for the safety net first

The reason to have an emergency fund instead of paying off the card: when the next emergency hits and you have no buffer, you put it on the same credit card you just paid off. And now you’re back where you started, only with a bunch of months of effort wasted and your psychological “I’m getting ahead” momentum gone.

The size of the safety net is a behavior bet, not a math bet. You’re betting that during the months it takes you to pay off the card, an unexpected expense will hit. Statistically, for most people, that’s a coin flip over a typical 18-month payoff horizon.

The answer most personal-finance people land on

A two-step structure that handles both:

  1. First, build a tiny emergency fund — $1,000. Not three months of expenses. Just $1,000. This is enough to absorb the most common single emergencies: a car repair, a medical copay, a flight home. Below $1,000, you’re not really safe from the credit-card-rebuild problem.

  2. Then attack the card aggressively until it’s at zero. Every dollar at this stage is earning 18%+ in avoided interest. Nothing else in your financial life has a higher guaranteed return.

  3. Then build the full emergency fund — 3 to 6 months of expenses. This is when you’re “really” safe.

The intuition behind step 1: $1,000 is small enough that you can build it in a month or two of focused saving, which means you’re not putting off card payoff for very long. But it’s big enough to break the loop where every emergency goes back on the card.

The numbers, with a real example

Say you have:

  • $4,500 on a card at 22% APR
  • $200/month of “extra” you can deploy (after minimums)
  • Current savings: $0

Path A: Save the full 3 months ($9,000) first, then attack the card

Months 1-45: All $200 goes to savings. Card balance grows from minimum payments not keeping up.

  • After 45 months of saving and paying minimums, savings = $9,000, card = still around $4,500 (minimums roughly equal interest at this point).
  • Then you attack the card with $200/month: ~26 more months.
  • Total: ~71 months. Interest paid: ~$5,200.

Path B: Card first, then save

Months 1-23: All $200 goes to the card. Paid off in ~23 months.

  • Then start saving $200/month for 45 months to reach $9,000.
  • Total: ~68 months. Interest paid: ~$1,100.

Path C: $1k buffer, then card, then full fund

Months 1-5: $200/month to savings. Buffer at $1,000 by month 5.

  • Months 6-28: $200/month to card. Paid off by month 28.
  • Months 29-69: $200/month to savings. Full $9,000 by month 69.
  • Total: ~69 months. Interest paid: ~$1,400.

Path C costs you about $300 more in interest than Path B, in exchange for never being one car repair away from re-using the card. For most people that’s a trade worth making.

When to deviate

  • Card APR below 8%: the math flips closer to “save first.” 8% credit card debt is rare (usually only from old credit-union cards or balance-transfer offers) but if you have it, prioritize the fund.
  • 0% promotional rates: just make minimum payments on the promo card, save aggressively, and pay it off before the promo expires. Treat the 0% balance like a free 12-18 month loan, not an emergency.
  • You can’t trust yourself not to re-use the card after payoff: cut up the card before paying it off. Then path A becomes less risky. (This is a behavior fix, not a math fix.)

The summary

For most people, in most situations, the right move is:

  1. $1k buffer
  2. Card to zero
  3. Full 3-6 month fund

The math says “card first” is mathematically best. The behavior says “tiny buffer first” is realistically best. The two-step compromise costs almost nothing and protects you from yourself.