Copper Ledger logo: a green ledger book with a copper coin Copper Ledger

Finance news with the numbers in view

Home / Personal Finance

One in Four Buy Now, Pay Later Users Paid Late, the Fed Finds. Why Holiday Surveys Show a Bigger Number

Federal Reserve data put late BNPL payments at 26% of users, while holiday polls show 38% expecting trouble. The two figures measure different things.

3100 Personal Finance 4 min read

Photo: Bautsch, CC0, via Wikimedia Commons / Wikimedia Commons, CC0

Two statistics about buy now, pay later are circulating this holiday season, and they sound like they contradict each other. A Federal Reserve survey says 26% of BNPL users paid late at least once in the past year. A September poll cited by 24/7 Wall St. says 38% of the shoppers who plan to use BNPL this holiday expect not to make every payment on time. Is the problem getting worse, or are the numbers simply different animals?

They are different animals. Sorting them out is more useful than picking one.

What the Fed measured

The central bank’s August issue of Consumer & Community Context draws on the 2025 Survey of Household Economics and Decisionmaking, a nationally representative survey of nearly 13,000 respondents. It reports that 16% of adults used BNPL in 2025. Among those users:

  • 26% paid late at least once in the prior year.
  • 17% were charged extra for paying late.
  • 11% had a BNPL payment trigger an overdraft or insufficient-funds fee.

The same publication stresses that “the vast majority of BNPL loans are repaid.” It also finds the overdraft problem is not spread evenly. The rate was 18% among users whose largest emergency expense they could cover from savings was under $100, and 4% among those who could cover $2,000 or more. Another line deserves attention: most BNPL lenders do not report loans to credit bureaus, so on-time payments generally earn no credit, though loans sent to collections do appear.

That is a look backward at what people did, across everyone who used the product.

What the holiday polls measured

The 38% figure comes from a NerdWallet and Harris Poll survey conducted September 8 to 10, as 24/7 Wall St. reports. It found 38% of holiday shoppers expect to use BNPL, and 38% of those planned users expect to miss a payment or pay late. That is a forecast made before the shopping, by people who already chose to say they would use the product, and the article gives no sample size.

The same article cites LendingTree, whose October 2 to 9 survey found 52% of Americans expecting to apply for BNPL within six months, up from 46% in September. And it quotes Block, which said consumer lending originations reached $19.0 billion in the third quarter, up 40% from a year earlier. Those are signs of demand, not of repayment trouble, and they should not be added to the late-payment numbers.

Lining the figures up shows why they cannot be compared directly:

  • The Fed counts what happened over a year to people who used BNPL, in a large probability survey.
  • The poll asks what people think will happen over one holiday season, among the subset who intend to use it, without a published sample size.
  • Neither figure, as reported, says how large the late payments were, how long they lasted, or how many were fixed within days.

The product is bigger than the “pay in 4” most people picture

A June Fed research note, “Buy Now, Pay Later Beyond ‘Pay in 4,’” estimates that six providers issued about $156.7 billion in U.S. BNPL credit in 2025. The six are Affirm, Afterpay, Klarna, PayPal, Sezzle and Zip. The authors put the familiar pay-in-four plans at about $78.3 billion, roughly half. Short-term installment loans came to about $31.2 billion and longer-term loans to about $47.1 billion. About 63% of total issuance carried a 0% APR, and the remaining 37% charged interest.

They estimate pay-in-four volume at nearly 80% above the Consumer Financial Protection Bureau’s 2023 figure of $43.9 billion. Two caveats come with that: the numbers are the authors’ estimates built from company filings, where U.S. data was missing they scaled global figures, and the note reflects the authors’ views, not necessarily the Federal Reserve Board’s.

This matters for the late-payment debate because the Fed’s August publication focuses on the original pay-in-four product. The longer loans, with interest and monthly schedules, are a different thing, and that publication says little about how they behave.

What remains unknown

Nobody in these sources has measured the gap between “expects to pay late” and “paid late” for the same people. That would take a follow-up survey after the season, and until one exists the holiday polls are best read as a mood indicator. The Fed’s 26% is a better anchor for what has already happened, and its finding that overdraft fees fall hardest on people with the thinnest savings is the clearest sign of who is most exposed.

The 24/7 Wall St. piece also points to weak consumer sentiment as the backdrop: the University of Michigan index stood at 48.1 in September, below the 60 level the article associates with recessions. Whether that gloom shows up in missed installments will be visible in lender data over the winter.

Sources

  1. Federal Reserve Board: What is Buy Now, Pay Later? (Consumer & Community Context, August 2026) federalreserve.gov
  2. Federal Reserve Board: Buy Now, Pay Later Beyond Pay in 4, a Comprehensive Product Overview (FEDS Notes, June 2026) federalreserve.gov
  3. 24/7 Wall St.: Americans Are Turning to Buy Now, Pay Later for the Holidays 247wallst.com