Every buy now, pay later checkout button comes with a little confetti animation and the phrase “no interest, ever.” That phrase is true often enough to make a great ad and rare enough to make a bad assumption. Roughly 96 million Americans are projected to use BNPL in 2026, up from about 91.5 million in 2025, and most of them get exactly the deal the checkout button promised on that one purchase. The costs show up later — when there’s a second loan stacked on the first, a payment gets missed, or the plan someone picked wasn’t actually the interest-free one.

What “Interest-Free” Actually Means

The claim is technically true for one specific product, not the whole company. Klarna, Afterpay, and Affirm all offer a “Pay in 4” style plan — four equal installments, no interest, no catch — and if that’s the plan someone signs up for and pays on schedule, the marketing checks out. The trap is that all three companies also sell longer monthly financing plans sitting right next to the interest-free option at checkout, and those plans charge real interest: Klarna’s monthly financing runs 7.99%–33.99% APR, Afterpay’s “Pay Monthly” option runs 6.99%–35.99% using simple (non-compounding) interest, and Affirm’s monthly plans span 0%–36% APR depending on the retailer and the shopper’s approval. “Buy now, pay later is interest-free” is a true sentence about a specific button. It is not a true sentence about the company that built the button.

In-Article Ad Slot — 300×250

The Real Cost Isn’t Interest — It’s Stacking

Even the interest-free Pay in 4 plans aren’t actually risk-free, because almost nobody uses just one. Industry data puts the share of BNPL users carrying multiple loans at the same time at 63%, and a third of users are juggling multiple providers — Klarna for one purchase, Afterpay for another, Affirm for a third — which means the four-tiny-payments math resets with every new tab. None of those apps can see what the others are doing, so nothing stops four “harmless” $15 biweekly payments from becoming one $60 biweekly obligation that never shows up as a single number anywhere. Missed-payment rates back this up: 34%–41% of BNPL users report missing at least one payment, even though full charge-offs (loans that go fully unpaid) stay low, around 1.8%–2%. The gap between those two numbers is the real story — most people eventually pay, but a lot of people are getting there on payment-friction fumes, and roughly 1 in 5 subprime BNPL users rack up an actual late fee.

The fee structures aren’t identical, either, which matters more once a plan is missed: Klarna charges up to $7 or 25% of the purchase price, whichever applies; Afterpay charges up to $8 or 25% of the order value; Affirm charges no late fees at all — a genuine structural difference, not a marketing one. Here’s what stacking a handful of ordinary purchases actually looks like on paper:

Purchase (illustrative) Provider Price Due every 2 weeks Late fee if missed
Sneakers Klarna $120 $30 up to $7 / 25%
Groceries splurge Afterpay $80 $20 up to $8 / 25%
Headphones Affirm (Pay in 4) $200 $50 $0
Home goods Klarna $60 $15 up to $7 / 25%
Total obligation $460 $115

Illustrative example built from representative purchase amounts, not a real transaction — the point is the arithmetic of stacking, not these specific items. Provider late-fee terms are accurate as of this writing.

The Regulatory Gap Nobody Mentions

Part of why BNPL gets pitched as basically consequence-free is that, for a while, federal regulators were on track to treat it like a credit card — with the dispute rights and refund protections that implies. In 2024, the Consumer Financial Protection Bureau issued an interpretive rule saying BNPL providers offering “Pay in 4” loans through digital accounts should be treated as credit card issuers under Regulation Z. That rule didn’t survive: the CFPB announced its intent to withdraw it in March 2025 and confirmed in a June 20, 2025 court filing that it will not reissue any revised version, citing both a procedural defect (it should have gone through full notice-and-comment rulemaking) and its own view that open-end credit card rules were a poor fit for short-term installment products. Practically, that means BNPL purchases carry fewer federally guaranteed dispute and refund protections than the “just like a credit card, but free” framing implies — and it’s part of why some states, including New York, are now moving to write their own BNPL licensing and consumer-protection rules instead of waiting on Washington. This is a fast-moving area of law; treat the regulatory picture above as accurate as of publication, not permanent.

Your Credit Score Is About to Notice

The other reason BNPL has felt consequence-free is that it mostly hasn’t touched credit scores — soft or no credit checks on approval, and loans that traditionally didn’t get reported the way a credit card balance does. That’s changing. FICO introduced two new scoring models, FICO Score 10 BNPL and FICO Score 10 T BNPL, in fall 2025, specifically built to fold BNPL loan data into the score at no extra cost to lenders. Adoption is optional, not mandatory — lenders can run the new scores side by side with the classic version rather than switching over immediately, so BNPL activity isn’t showing up in every score pulled today. But the direction is clear: the years when stacking four “pay in 4” plans felt invisible to your credit file are ending, unevenly, plan by plan, lender by lender. Treating BNPL like the debt it is — not like a loophole — stops being optional right around the same time it stops being invisible.