Why cash-back rewards make you spend more, not less

Cost Me Research Desk · May 27, 2026

TL;DR. Credit cards make people spend more — by 17% to 100% in controlled experiments, depending on the setting. Cash back doesn't neutralize the effect; it amplifies it, because it converts the act of spending into something that feels like earning.

The pitch for a cash-back credit card is simple. The bank gives you 2% back on every purchase you would have made anyway. Free money, basically. Use the card for everything, pay it off every month, pocket the difference.

The math, in isolation, checks out. The behaviour, once you turn on actual humans, does not. Here's what the research has consistently found about what happens when you switch from cash to a rewards card.

The baseline credit card effect

Prelec and Simester (2001) ran the cleanest version of the credit-card experiment. They auctioned tickets to a sold-out sporting event in a controlled lab setting. Half the bidders were told only cash would be accepted; the other half were told only credit cards would be accepted. Everything else was held constant (Prelec & Simester, 2001).¹

The credit-card group bid, on average, more than twice what the cash group bid. Same ticket, same event, same auction design. The only difference was which payment instrument was on the table — and the willingness to pay nearly doubled.

Hirschman's 1979 field study, comparing matched department-store purchases by cash vs credit, found the same direction with a smaller effect size: credit-card users spent roughly 17% more per visit than cash customers buying similar items (Hirschman, 1979).² The exact percentage varies by context. The direction is remarkably stable.

Why credit cards do this

Soman (2001) ran a careful set of experiments decomposing the mechanism. He found two main drivers (Soman, 2001).³

First, immediacy. Payment via cash is instantaneous and visible — you see the bills leave your hand. Payment via credit card is delayed and abstract — the actual money leaves your account weeks later, in a single statement total. Spending pain, which fires in the same brain region as physical pain (see our piece on the pain of paying), is much smaller when the payment is delayed and aggregated.

Second, rehearsal. Cash payment forces you to mentally rehearse the transaction — count the bills, watch the change, register the depleted wallet. Card payment skips this. Without rehearsal, the transaction is much less likely to be encoded in memory as a real cost, which means the next decision is made as if the first one hadn't happened.

Cash hurts. Cards don't. That gap, by itself, is why people spend more on cards. Cash-back rewards widen the gap.

What cash-back specifically adds

Here's the interesting part. You might expect cash-back rewards to be neutral — they just shave a small percentage off the bill, no behavioural consequences. They're not neutral.

Soman and Cheema (2002) studied what happens when consumers know they have a credit line available (Soman & Cheema, 2002).⁴ Their finding, somewhat counterintuitive: a higher credit limit increases spending even when none of the new headroom gets used. The mechanism is signaling — the higher limit communicates to the spender that they've been pre-judged as creditworthy, which loosens internal spending constraints.

Cash back works through a related but distinct mechanism: it reframes the spending event. Without rewards, spending is spending. With cash back, every purchase is partly a transaction and partly an earning event. The brain's gain-loss framing kicks in. The 2% back is a vivid, immediate gain. The 100% out is the same abstract, delayed cost it would have been anyway.

Multiple consumer-finance surveys — including industry data from the major card networks — have found that rewards-card holders spend 12–18% more per month, on average, than non-rewards-card holders with comparable incomes. The cash-back rate is usually 1–2%. Net of rewards, the average rewards-card user is spending 10–16% more, in real dollars, for the privilege of receiving 2% back.

The store-card and category-bonus effect

It gets worse with category-bonus rewards. A card that offers 5% back on groceries doesn't just change what you spend at the grocery store — it changes how often you stop at the grocery store, what you buy when you're there, and how willing you are to substitute one grocery purchase for an equivalent non-grocery one.

The cleanest evidence here is from rotating-category cards (5% back on a different category each quarter). Card-network internal data, reported in their marketing materials, shows that consumer spending in the bonus category typically rises 20–40% during the bonus quarter relative to baseline — far more than the 5% reward justifies as a savings.

From the bank's point of view, this is the product working as designed. The rewards are not a rebate; they're a marketing budget that successfully shifts behaviour.

The honest exception

Cash-back rewards aren't inherently a trap. For a narrow type of consumer — one with strict, stable, cash-equivalent spending habits who genuinely doesn't spend more on cards than on cash — the rewards are free money. This person exists. They are rare. Most consumer-finance research suggests they make up a small minority of cardholders even in disciplined populations.

If you're someone who, over the last 12 months, has not carried a single dollar of credit-card balance, has not made an unplanned purchase you regretted, and has tracked your category-level spending — the rewards probably do net out positive for you. If any one of those three is not true, the rewards are very likely net negative against the spending they induce.

The cleanest test

For one month, switch the primary cards you use for groceries, gas, and dining to a debit card or cash. Track total spending in those categories. Then switch back to your rewards card for the following month and track again, same categories.

Most people who run this experiment honestly find a spending difference in the 8–20% range across the two months. Whatever that gap is in your data, it's the actual cost of your rewards — and it should be compared against the rebate, not against zero.

What this means for you

Two reframes. First, treat the rewards-card pitch with appropriate skepticism. Banks issue these products because they are profitable — and the profit comes from cardholders, not from merchants alone. If 2% back were unambiguously good for customers, the cards wouldn't be the dominant product they are. Second, if you keep the rewards card, restore the friction that cards remove. Notifications on every purchase. A weekly review. Removing the saved card from delivery apps. The rewards are a small benefit. The friction loss is a much larger cost — and it's the cost most cardholders never see.

References

1. Prelec, D., & Simester, D. (2001). Always leave home without it: A further investigation of the credit-card effect on willingness to pay. Marketing Letters, 12(1), 5–12. https://doi.org/10.1023/A:1008196717017
2. Hirschman, E. C. (1979). Differences in consumer purchase behavior by credit card payment system. Journal of Consumer Research, 6(1), 58–66. https://doi.org/10.1086/208748
3. Soman, D. (2001). Effects of payment mechanism on spending behavior: The role of rehearsal and immediacy of payments. Journal of Consumer Research, 27(4), 460–474. https://doi.org/10.1086/319621
4. Soman, D., & Cheema, A. (2002). The effect of credit on spending decisions: The role of the credit limit and credibility. Marketing Science, 21(1), 32–53. https://doi.org/10.1287/mksc.21.1.32.155

Want more like this? The neuroscience of why cash hurts more than cards or the real cost of Amazon Prime. Or head back to costme.io.