Blog

Three households use one cash back card. Who actually comes out ahead?

Invented households run the same rewards math with different habits: paying in full, carrying a balance, paying a fee on low spending, and spending to earn.

A cash back card does not have a value of its own. It has a value to a particular household, and that value can be positive, near zero or negative depending on how the card is used. The rewards rate on the ad is the same for everyone; the interest, fees and extra spending around it are not.

This post follows three households, plus one short extra case, through a year of the same invented card. Every household is hypothetical, and every rate, fee and balance is a round number chosen to make the arithmetic checkable. They are illustrations, not a survey of how people behave and not any real card's terms. The point is to give you a way to run the same tally on your own year.

The net cash back formula

Before the households, one formula. It is the tally to run on any rewards card:

net result = rewards earned − interest paid − fees paid − spending you would not have done without the card

Most people count only the first term. The other three are where a good deal turns bad. A reward is a percentage of spending, so it grows slowly with spending. Interest is a percentage of a balance, charged month after month, and for a typical balance it is much larger than a point or two of rewards. Fees are fixed dollars that do not care how much you spent.

The card in the examples is simple. Card X is a flat-rate card paying 2 percent cash back on every purchase with no annual fee, and an illustrative 24 percent APR. For comparison, the Federal Reserve's G.19 release put the average APR on card accounts assessed interest at 22.36 percent in the third quarter of 2026, so the 24 percent used here is in a realistic range without being anyone's actual rate.

Household 1: Priya and Marcus pay in full, every month

Priya and Marcus put their groceries, utilities, fuel and most other spending on Card X: $3,500 a month, or $42,000 a year. They set up autopay for the full statement balance, so they never carry interest.

  • Rewards: $42,000 × 2% = $840.
  • Interest: $0, because the balance is paid in full each month and the grace period on new purchases is preserved.
  • Fees: $0.
  • Extra spending because of the card: none; they were going to buy all of this anyway.

Net result: $840 for the year.

They come out clearly ahead, and the reason is simple: the only term that is not zero is the reward. The household is the group card marketing is aimed at. The thing that would hurt them is a slip. If one autopay failed and a payment arrived late, the issuer could charge a late fee of roughly $30 for a first occurrence, and, because the statement balance was not paid in full by the due date, interest on new purchases could begin immediately. A single $30 fee wipes out the rewards on $30 ÷ 0.02 = $1,500 of spending. That is not a reason to avoid the card. It is a reason to keep autopay on, to watch the account for the notice that a payment failed and to call the issuer's number on the back of the card if one does.

Household 2: Lena earns rewards and pays more in interest

Lena spends $1,800 a month, $21,600 a year, on Card X. A year ago a medical bill left her with a balance she has not cleared. Across the year she carries an average balance of $2,400. She makes at least the minimum each month and is never late.

  • Rewards: $21,600 × 2% = $432.
  • Interest: $2,400 × 24% = $576, a simple estimate that ignores compounding and payment timing.
  • Fees: $0.
  • Extra spending: none.

Net result: $432 − $576 = −$144.

Lena's net is negative even though she earns more than $400 in rewards. There is also something the estimate understates. In the usual design a card's grace period on new purchases applies only if the previous statement balance was paid in full. Lena does not pay in full, so her new purchases may accrue interest from the day they post, and the real interest bill is a little higher than $576. The card's agreement states the exact method.

The break-even APR

You can turn this into a single number. Lena's rewards divided by her average balance is $432 ÷ $2,400 = 0.18, or 18 percent. As long as her APR is above 18 percent, the interest costs more than the cash back. At 24 percent she loses; at the Fed's 22.36 percent average she would still lose; only a rate below 18 percent would put her ahead. For another household the break-even APR is the reward earned in a year divided by the average balance carried in the same year. If your APR is above it, rewards are not paying for the balance.

Lena's year, month by month

The yearly figures hide how the loss arrives, so here is the monthly view. With a $2,400 balance and a 24 percent APR, the monthly periodic rate is about 24% ÷ 12 = 2 percent, so each month's interest is $2,400 × 2% = $48. Her monthly rewards are $1,800 × 2% = $36. Each month she is $12 behind, and $12 × 12 = $144, which matches the yearly figure.

Notice that the rate is the same 2 percent in both places, a coincidence that is worth remembering. A 24 percent APR charges about 2 percent a month on a balance. A 2 percent reward pays 2 percent once on a purchase. Carrying a purchase for a single month at that rate gives back to the issuer the entire reward the purchase earned.

What a payoff would look like

Say Lena decides to attack the old balance. To keep the illustration clean, suppose she stops adding to it and sends $200 a month toward the $2,400, with interest of 2 percent a month on what is left. Month by month, the interest is $48.00, then $44.96, then $41.86, and it keeps shrinking as the balance falls. In the fourteenth month the balance is nearly gone and the last payment is about $171.97. Adding it up, she makes 13 payments of $200 and a final one of $171.97, about $2,771.97 in total, of which $371.97 is interest.

That is 14 months and under $400 of interest. Compare it with the $576 she would pay in just one more year of carrying the balance. The numbers are simplifications: they ignore any new purchases that would accrue interest, they assume a fixed rate, and a real statement rounds differently. Her own statement and the credit card payoff calculator will give her the real schedule.

What Lena can do

None of this means Lena did something wrong. A medical bill is not a spending choice. It means the problem to solve is the balance, not the rewards rate:

  1. Find the real cost. Her statement shows the interest charged each month, and the credit card payoff calculator turns a balance, rate and payment into a payoff date.
  2. Ask about lower-cost options. A call to the number on the back of the card can open a hardship or reduced-rate program in some cases, and how to ask your card issuer for a hardship program, call by call covers what to ask.
  3. Consider moving the balance to a lower rate for a period, if the fee and the deadline make arithmetic sense. a balance transfer from start to finish, fee and deadline included shows how to test that.
  4. Get a payoff plan. A nonprofit credit counseling agency can review the whole picture, and how to build a credit card payoff plan you can actually stick to sets out a plan that starts with the minimums.

If the issuer ever mishandles the account, the CFPB complaint process is open to anyone.

Household 3: Omar pays a fee on small spending

Omar is careful with money. He spends $700 a month on a card, $8,400 a year, and pays the balance in full. He chose a card that looks richer than Card X: 3 percent on groceries and dining, 1 percent on everything else, and a $95 annual fee. Of his $700, $300 a month is groceries and dining and $400 is everything else.

  • Rewards on groceries and dining: $300 × 12 = $3,600, times 3% = $108.
  • Rewards on everything else: $400 × 12 = $4,800, times 1% = $48.
  • Total rewards: $156.
  • Annual fee: $95.

Net result: $156 − $95 = $61.

Now compare it with Card X on the same spending: $8,400 × 2% = $168, with no fee. Omar's fee card leaves him $168 − $61 = $107 a year behind the card he did not choose.

The mechanism is the fee. A fixed fee is paid whether Omar spends $700 or $7,000, and the extra 1 percent on a bonus category is worth only $36 a year on his $3,600 of groceries and dining ($3,600 × 1% = $36). He pays $95 for a $36 advantage, then earns 1 percent on the rest where the flat card earns 2 percent, which costs him $48 on that $4,800. The fee card does not make sense for a household with this little spending and this much of it outside the bonus categories. A fee card needs to be tested like this before it is opened, and should a cash-back card with an annual fee ever beat a no-fee one? provides the decision steps.

A fourth case: Taylor spends more to earn more

Taylor is the shortest case and the one that gets overlooked. Taylor already spends $1,500 a month on Card X. Because the card pays back, Taylor starts using it for a few extra purchases, an upgrade here and a sale there, adding $300 a month that would not otherwise have been spent.

  • Extra spending: $300 × 12 = $3,600.
  • Rewards on the extra spending: $3,600 × 2% = $72.

If the purchases were not needed, Taylor's net result on this slice is $72 − $3,600 = −$3,528. Of every $100 spent only to earn rewards, $2 comes back and $98 is gone.

Not every extra purchase is wasted. A reasonable reader could argue that some of Taylor's extra spending was useful. The question to ask is the one in the formula: would I have bought this without the card? If the answer is no, the reward is not a discount, because nothing was being bought at a price you wanted to pay. spending more to earn more: how chasing cash back costs you money goes through how this happens.

The households side by side

Household Annual spending Rewards Interest Fee Net on the formula
Priya and Marcus $42,000 $840 $0 $0 +$840
Lena $21,600 $432 $576 $0 −$144
Omar (fee card) $8,400 $156 $0 $95 +$61
Omar (Card X instead) $8,400 $168 $0 $0 +$168
Taylor (extra slice only) $3,600 $72 $0 $0 −$3,528 if not needed

The table is not a ranking of people. It is a ranking of habits. Priya and Marcus win by paying in full. Lena loses to interest despite earning more than Omar does. Omar leaves money on the table by choosing the wrong design for his level of spending. Taylor's loss has nothing to do with the card.

What each household would change

A tally is only useful if it points to something you would do differently. For the four cases, the changes are small and different:

  • Priya and Marcus do not need a better card. They need to keep what is working: autopay on for the full statement balance, an alert for failed payments and a habit of reading the statement when it arrives. If they compare cards at all, the gain from a rate of 2 percent instead of 1.5 percent on $42,000 is $840 − $630 = $210 a year, real money but small beside what an avoidable slip could cost.
  • Lena needs to change the balance, not the rewards. The same $432 of rewards is worth less than the $576 she pays in interest, and every month of carrying the balance makes the gap larger.
  • Omar needs to either switch to a no-fee flat card or find out whether his spending in the bonus categories is far larger than he thinks. At $300 a month of bonus spending, the fee does not pay.
  • Taylor needs a rule, not a card. Before any purchase made partly for the reward, ask whether you would buy it at this price without the reward.

What the tally leaves out

The formula is deliberately narrow, and a few things sit outside it. None of them changes the lesson of the examples, but they belong in a complete picture.

  • Redemption value. The examples assume a dollar of cash back is worth a dollar. Many cards pay cash back at face value, but a redemption as a gift card or a certificate can be worth less, and a rule that sets a minimum redemption amount can leave a small balance unspent for months. The card's terms spell this out, and seven checks to run before you redeem credit card rewards lists what to look at.
  • Card benefits. Purchase protection, extended warranty and return protection are issuer or network benefits, not federal law, and the card's guide to benefits sets their limits. They can have real value that this tally does not count, and they can also be narrower than people expect.
  • Credit effects. A card that is used lightly and paid in full is generally good for a credit file, since utilization stays low and the account ages. A card that carries a high balance relative to its limit can do the reverse. That cost never appears on the rewards tally and is hard to price.
  • Opportunity cost. If Priya and Marcus would also qualify for a lower rate somewhere else or earn more on a different card, the comparison changes, but only if the other option is real for them.

The simplest way to use the tally is as a first filter. If the net is positive and the habits behind it are stable, the rewards are a bonus. If the net is negative, fix the term causing it, whether interest, fee or extra spending, before worrying about which card earns the most.

Find your own row

Take your last twelve statements and work through these steps:

  1. Add up the year's spending on the card. Use the annual total on your December statement or sum the months.
  2. Multiply by your reward rate, or work through each category if the card pays different rates.
  3. Add up the interest charged. It is on each statement. If the total is above zero, find the average balance and compute your break-even APR as rewards divided by balance.
  4. Add up the fees, including the annual fee, late fees and any foreign transaction fees.
  5. Subtract. If the answer is negative, rewards are not the problem to solve.
  6. Ask the extra-spending question about the last big purchase made with the card in mind.

Key takeaways

  • A card's value is rewards minus interest, fees and spending you would not have done. Run all four terms, not just the first.
  • Paying the full statement balance each month is what lets rewards count as a gain. In the example the rewards were worth $840 a year to a household that never paid interest.
  • The break-even APR is annual rewards divided by average balance. In Lena's case it was 18 percent, below the Fed's reported 22.36 percent average for accounts that are charged interest, so rewards could not cover her interest.
  • A fee card can lose to a no-fee flat card at low spending. Omar's $95 fee turned a $156 reward into a $61 result against $168 on the simple card.
  • Spending more to earn rewards returns about two cents on the dollar. Rewards reward what you were going to buy anyway.

Terms used in this article

Related tools

Latest on this topic